API tools faq
paste
Guest User

Untitled

a guest
Mar 1st, 2025
60
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 68.98 KB | None | 0 0
raw download clone embed print report
  1. 2025/03/01 19:54:31 routes.go:1205: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[chrome-extension://eobbhoofkanlmddnplfhnmkfbnlhpbbo http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
  2. time=2025-03-01T19:54:31.712Z level=INFO source=images.go:432 msg="total blobs: 43"
  3. time=2025-03-01T19:54:31.720Z level=INFO source=images.go:439 msg="total unused blobs removed: 0"
  4. time=2025-03-01T19:54:31.729Z level=INFO source=routes.go:1256 msg="Listening on [::]:11434 (version 0.5.12)"
  5. time=2025-03-01T19:54:31.729Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
  6. time=2025-03-01T19:54:31.882Z level=INFO source=types.go:130 msg="inference compute" id=GPU-b43abc1b-15ae-ef30-9262-e096cf33a0ec library=cuda variant=v12 compute=8.9 driver=12.4 name="NVIDIA GeForce RTX 4090" total="23.6 GiB" available="23.3 GiB"
  7. [GIN] 2025/03/01 - 19:55:01 | 200 | 51.653µs | 192.168.17.19 | GET "/"
  8. [GIN] 2025/03/01 - 19:55:06 | 200 | 25.814µs | 192.168.17.19 | GET "/"
  9. [GIN] 2025/03/01 - 19:55:11 | 200 | 48.295µs | 192.168.17.19 | GET "/"
  10. [GIN] 2025/03/01 - 19:55:16 | 200 | 56.258µs | 192.168.17.19 | GET "/"
  11. [GIN] 2025/03/01 - 19:55:21 | 200 | 32.518µs | 192.168.17.19 | GET "/"
  12. [GIN] 2025/03/01 - 19:55:26 | 200 | 30.648µs | 192.168.17.19 | GET "/"
  13. time=2025-03-01T19:55:28.941Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.head_count_kv default=1
  14. time=2025-03-01T19:55:28.941Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.key_length default=64
  15. time=2025-03-01T19:55:28.941Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.value_length default=64
  16. time=2025-03-01T19:55:28.941Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.head_count_kv default=1
  17. time=2025-03-01T19:55:28.941Z level=INFO source=sched.go:715 msg="new model will fit in available VRAM in single GPU, loading" model=/root/.ollama/models/blobs/sha256-819c2adf5ce6df2b6bd2ae4ca90d2a69f060afeb438d0c171db57daa02e39c3d gpu=GPU-b43abc1b-15ae-ef30-9262-e096cf33a0ec parallel=1 available=24973475840 required="1.1 GiB"
  18. time=2025-03-01T19:55:29.045Z level=INFO source=server.go:97 msg="system memory" total="62.8 GiB" free="57.9 GiB" free_swap="0 B"
  19. time=2025-03-01T19:55:29.045Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.head_count_kv default=1
  20. time=2025-03-01T19:55:29.045Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.key_length default=64
  21. time=2025-03-01T19:55:29.045Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.value_length default=64
  22. time=2025-03-01T19:55:29.045Z level=WARN source=ggml.go:132 msg="key not found" key=bert.attention.head_count_kv default=1
  23. time=2025-03-01T19:55:29.045Z level=INFO source=server.go:130 msg=offload library=cuda layers.requested=-1 layers.model=25 layers.offload=25 layers.split="" memory.available="[23.3 GiB]" memory.gpu_overhead="0 B" memory.required.full="1.1 GiB" memory.required.partial="1.1 GiB" memory.required.kv="3.0 MiB" memory.required.allocations="[1.1 GiB]" memory.weights.total="580.2 MiB" memory.weights.repeating="520.6 MiB" memory.weights.nonrepeating="59.6 MiB" memory.graph.full="8.0 MiB" memory.graph.partial="8.0 MiB"
  24. time=2025-03-01T19:55:29.060Z level=INFO source=server.go:380 msg="starting llama server" cmd="/usr/bin/ollama runner --model /root/.ollama/models/blobs/sha256-819c2adf5ce6df2b6bd2ae4ca90d2a69f060afeb438d0c171db57daa02e39c3d --ctx-size 512 --batch-size 512 --n-gpu-layers 25 --threads 6 --parallel 1 --port 42893"
  25. time=2025-03-01T19:55:29.061Z level=INFO source=sched.go:450 msg="loaded runners" count=1
  26. time=2025-03-01T19:55:29.061Z level=INFO source=server.go:557 msg="waiting for llama runner to start responding"
  27. time=2025-03-01T19:55:29.061Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error"
  28. time=2025-03-01T19:55:29.093Z level=INFO source=runner.go:932 msg="starting go runner"
  29. ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
  30. ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
  31. ggml_cuda_init: found 1 CUDA devices:
  32. Device 0: NVIDIA GeForce RTX 4090, compute capability 8.9, VMM: yes
  33. load_backend: loaded CUDA backend from /usr/lib/ollama/cuda_v12/libggml-cuda.so
  34. load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-sandybridge.so
  35. time=2025-03-01T19:55:29.152Z level=INFO source=runner.go:935 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | CUDA : ARCHS = 600,610,620,700,720,750,800,860,870,890,900 | USE_GRAPHS = 1 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | LLAMAFILE = 1 | cgo(gcc)" threads=6
  36. time=2025-03-01T19:55:29.153Z level=INFO source=runner.go:993 msg="Server listening on 127.0.0.1:42893"
  37. llama_load_model_from_file: using device CUDA0 (NVIDIA GeForce RTX 4090) - 23816 MiB free
  38. llama_model_loader: loaded meta data with 23 key-value pairs and 389 tensors from /root/.ollama/models/blobs/sha256-819c2adf5ce6df2b6bd2ae4ca90d2a69f060afeb438d0c171db57daa02e39c3d (version GGUF V3 (latest))
  39. llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
  40. llama_model_loader: - kv 0: general.architecture str = bert
  41. llama_model_loader: - kv 1: general.name str = mxbai-embed-large-v1
  42. llama_model_loader: - kv 2: bert.block_count u32 = 24
  43. llama_model_loader: - kv 3: bert.context_length u32 = 512
  44. llama_model_loader: - kv 4: bert.embedding_length u32 = 1024
  45. llama_model_loader: - kv 5: bert.feed_forward_length u32 = 4096
  46. llama_model_loader: - kv 6: bert.attention.head_count u32 = 16
  47. llama_model_loader: - kv 7: bert.attention.layer_norm_epsilon f32 = 0.000000
  48. llama_model_loader: - kv 8: general.file_type u32 = 1
  49. llama_model_loader: - kv 9: bert.attention.causal bool = false
  50. llama_model_loader: - kv 10: bert.pooling_type u32 = 2
  51. llama_model_loader: - kv 11: tokenizer.ggml.token_type_count u32 = 2
  52. llama_model_loader: - kv 12: tokenizer.ggml.bos_token_id u32 = 101
  53. llama_model_loader: - kv 13: tokenizer.ggml.eos_token_id u32 = 102
  54. llama_model_loader: - kv 14: tokenizer.ggml.model str = bert
  55. llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,30522] = ["[PAD]", "[unused0]", "[unused1]", "...
  56. llama_model_loader: - kv 16: tokenizer.ggml.scores arr[f32,30522] = [-1000.000000, -1000.000000, -1000.00...
  57. llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,30522] = [3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
  58. llama_model_loader: - kv 18: tokenizer.ggml.unknown_token_id u32 = 100
  59. llama_model_loader: - kv 19: tokenizer.ggml.seperator_token_id u32 = 102
  60. llama_model_loader: - kv 20: tokenizer.ggml.padding_token_id u32 = 0
  61. llama_model_loader: - kv 21: tokenizer.ggml.cls_token_id u32 = 101
  62. llama_model_loader: - kv 22: tokenizer.ggml.mask_token_id u32 = 103
  63. llama_model_loader: - type f32: 243 tensors
  64. llama_model_loader: - type f16: 146 tensors
  65. time=2025-03-01T19:55:29.312Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model"
  66. llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
  67. llm_load_vocab: special tokens cache size = 5
  68. llm_load_vocab: token to piece cache size = 0.2032 MB
  69. llm_load_print_meta: format = GGUF V3 (latest)
  70. llm_load_print_meta: arch = bert
  71. llm_load_print_meta: vocab type = WPM
  72. llm_load_print_meta: n_vocab = 30522
  73. llm_load_print_meta: n_merges = 0
  74. llm_load_print_meta: vocab_only = 0
  75. llm_load_print_meta: n_ctx_train = 512
  76. llm_load_print_meta: n_embd = 1024
  77. llm_load_print_meta: n_layer = 24
  78. llm_load_print_meta: n_head = 16
  79. llm_load_print_meta: n_head_kv = 16
  80. llm_load_print_meta: n_rot = 64
  81. llm_load_print_meta: n_swa = 0
  82. llm_load_print_meta: n_embd_head_k = 64
  83. llm_load_print_meta: n_embd_head_v = 64
  84. llm_load_print_meta: n_gqa = 1
  85. llm_load_print_meta: n_embd_k_gqa = 1024
  86. llm_load_print_meta: n_embd_v_gqa = 1024
  87. llm_load_print_meta: f_norm_eps = 1.0e-12
  88. llm_load_print_meta: f_norm_rms_eps = 0.0e+00
  89. llm_load_print_meta: f_clamp_kqv = 0.0e+00
  90. llm_load_print_meta: f_max_alibi_bias = 0.0e+00
  91. llm_load_print_meta: f_logit_scale = 0.0e+00
  92. llm_load_print_meta: n_ff = 4096
  93. llm_load_print_meta: n_expert = 0
  94. llm_load_print_meta: n_expert_used = 0
  95. llm_load_print_meta: causal attn = 0
  96. llm_load_print_meta: pooling type = 2
  97. llm_load_print_meta: rope type = 2
  98. llm_load_print_meta: rope scaling = linear
  99. llm_load_print_meta: freq_base_train = 10000.0
  100. llm_load_print_meta: freq_scale_train = 1
  101. llm_load_print_meta: n_ctx_orig_yarn = 512
  102. llm_load_print_meta: rope_finetuned = unknown
  103. llm_load_print_meta: ssm_d_conv = 0
  104. llm_load_print_meta: ssm_d_inner = 0
  105. llm_load_print_meta: ssm_d_state = 0
  106. llm_load_print_meta: ssm_dt_rank = 0
  107. llm_load_print_meta: ssm_dt_b_c_rms = 0
  108. llm_load_print_meta: model type = 335M
  109. llm_load_print_meta: model ftype = F16
  110. llm_load_print_meta: model params = 334.09 M
  111. llm_load_print_meta: model size = 637.85 MiB (16.02 BPW)
  112. llm_load_print_meta: general.name = mxbai-embed-large-v1
  113. llm_load_print_meta: BOS token = 101 '[CLS]'
  114. llm_load_print_meta: EOS token = 102 '[SEP]'
  115. llm_load_print_meta: UNK token = 100 '[UNK]'
  116. llm_load_print_meta: SEP token = 102 '[SEP]'
  117. llm_load_print_meta: PAD token = 0 '[PAD]'
  118. llm_load_print_meta: CLS token = 101 '[CLS]'
  119. llm_load_print_meta: MASK token = 103 '[MASK]'
  120. llm_load_print_meta: LF token = 0 '[PAD]'
  121. llm_load_print_meta: EOG token = 102 '[SEP]'
  122. llm_load_print_meta: max token length = 21
  123. llm_load_tensors: offloading 24 repeating layers to GPU
  124. llm_load_tensors: offloading output layer to GPU
  125. llm_load_tensors: offloaded 25/25 layers to GPU
  126. llm_load_tensors: CUDA0 model buffer size = 577.22 MiB
  127. llm_load_tensors: CPU_Mapped model buffer size = 60.63 MiB
  128. llama_new_context_with_model: n_seq_max = 1
  129. llama_new_context_with_model: n_ctx = 512
  130. llama_new_context_with_model: n_ctx_per_seq = 512
  131. llama_new_context_with_model: n_batch = 512
  132. llama_new_context_with_model: n_ubatch = 512
  133. llama_new_context_with_model: flash_attn = 0
  134. llama_new_context_with_model: freq_base = 10000.0
  135. llama_new_context_with_model: freq_scale = 1
  136. llama_kv_cache_init: kv_size = 512, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 24, can_shift = 1
  137. llama_kv_cache_init: CUDA0 KV buffer size = 48.00 MiB
  138. llama_new_context_with_model: KV self size = 48.00 MiB, K (f16): 24.00 MiB, V (f16): 24.00 MiB
  139. llama_new_context_with_model: CUDA_Host output buffer size = 0.00 MiB
  140. llama_new_context_with_model: CUDA0 compute buffer size = 25.01 MiB
  141. llama_new_context_with_model: CUDA_Host compute buffer size = 5.01 MiB
  142. llama_new_context_with_model: graph nodes = 849
  143. llama_new_context_with_model: graph splits = 4 (with bs=512), 2 (with bs=1)
  144. time=2025-03-01T19:55:29.563Z level=INFO source=server.go:596 msg="llama runner started in 0.50 seconds"
  145. llama_model_loader: loaded meta data with 23 key-value pairs and 389 tensors from /root/.ollama/models/blobs/sha256-819c2adf5ce6df2b6bd2ae4ca90d2a69f060afeb438d0c171db57daa02e39c3d (version GGUF V3 (latest))
  146. llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
  147. llama_model_loader: - kv 0: general.architecture str = bert
  148. llama_model_loader: - kv 1: general.name str = mxbai-embed-large-v1
  149. llama_model_loader: - kv 2: bert.block_count u32 = 24
  150. llama_model_loader: - kv 3: bert.context_length u32 = 512
  151. llama_model_loader: - kv 4: bert.embedding_length u32 = 1024
  152. llama_model_loader: - kv 5: bert.feed_forward_length u32 = 4096
  153. llama_model_loader: - kv 6: bert.attention.head_count u32 = 16
  154. llama_model_loader: - kv 7: bert.attention.layer_norm_epsilon f32 = 0.000000
  155. llama_model_loader: - kv 8: general.file_type u32 = 1
  156. llama_model_loader: - kv 9: bert.attention.causal bool = false
  157. llama_model_loader: - kv 10: bert.pooling_type u32 = 2
  158. llama_model_loader: - kv 11: tokenizer.ggml.token_type_count u32 = 2
  159. llama_model_loader: - kv 12: tokenizer.ggml.bos_token_id u32 = 101
  160. llama_model_loader: - kv 13: tokenizer.ggml.eos_token_id u32 = 102
  161. llama_model_loader: - kv 14: tokenizer.ggml.model str = bert
  162. llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,30522] = ["[PAD]", "[unused0]", "[unused1]", "...
  163. llama_model_loader: - kv 16: tokenizer.ggml.scores arr[f32,30522] = [-1000.000000, -1000.000000, -1000.00...
  164. llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,30522] = [3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
  165. llama_model_loader: - kv 18: tokenizer.ggml.unknown_token_id u32 = 100
  166. llama_model_loader: - kv 19: tokenizer.ggml.seperator_token_id u32 = 102
  167. llama_model_loader: - kv 20: tokenizer.ggml.padding_token_id u32 = 0
  168. llama_model_loader: - kv 21: tokenizer.ggml.cls_token_id u32 = 101
  169. llama_model_loader: - kv 22: tokenizer.ggml.mask_token_id u32 = 103
  170. llama_model_loader: - type f32: 243 tensors
  171. llama_model_loader: - type f16: 146 tensors
  172. llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
  173. llm_load_vocab: special tokens cache size = 5
  174. llm_load_vocab: token to piece cache size = 0.2032 MB
  175. llm_load_print_meta: format = GGUF V3 (latest)
  176. llm_load_print_meta: arch = bert
  177. llm_load_print_meta: vocab type = WPM
  178. llm_load_print_meta: n_vocab = 30522
  179. llm_load_print_meta: n_merges = 0
  180. llm_load_print_meta: vocab_only = 1
  181. llm_load_print_meta: model type = ?B
  182. llm_load_print_meta: model ftype = all F32
  183. llm_load_print_meta: model params = 334.09 M
  184. llm_load_print_meta: model size = 637.85 MiB (16.02 BPW)
  185. llm_load_print_meta: general.name = mxbai-embed-large-v1
  186. llm_load_print_meta: BOS token = 101 '[CLS]'
  187. llm_load_print_meta: EOS token = 102 '[SEP]'
  188. llm_load_print_meta: UNK token = 100 '[UNK]'
  189. llm_load_print_meta: SEP token = 102 '[SEP]'
  190. llm_load_print_meta: PAD token = 0 '[PAD]'
  191. llm_load_print_meta: CLS token = 101 '[CLS]'
  192. llm_load_print_meta: MASK token = 103 '[MASK]'
  193. llm_load_print_meta: LF token = 0 '[PAD]'
  194. llm_load_print_meta: EOG token = 102 '[SEP]'
  195. llm_load_print_meta: max token length = 21
  196. llama_model_load: vocab only - skipping tensors
  197. [GIN] 2025/03/01 - 19:55:29 | 200 | 1.073499699s | 10.244.243.205 | POST "/api/embed"
  198. time=2025-03-01T19:55:30.714Z level=INFO source=sched.go:508 msg="updated VRAM based on existing loaded models" gpu=GPU-b43abc1b-15ae-ef30-9262-e096cf33a0ec library=cuda total="23.6 GiB" available="22.2 GiB"
  199. time=2025-03-01T19:55:30.716Z level=WARN source=ggml.go:132 msg="key not found" key=granite.attention.key_length default=64
  200. time=2025-03-01T19:55:30.716Z level=WARN source=ggml.go:132 msg="key not found" key=granite.attention.value_length default=64
  201. time=2025-03-01T19:55:30.716Z level=INFO source=sched.go:715 msg="new model will fit in available VRAM in single GPU, loading" model=/root/.ollama/models/blobs/sha256-1aefcd9a8a15091b670951963b5f8a7e6653bb1350345e9621e179685ac9bc5f gpu=GPU-b43abc1b-15ae-ef30-9262-e096cf33a0ec parallel=4 available=23866638336 required="11.2 GiB"
  202. time=2025-03-01T19:55:30.798Z level=INFO source=server.go:97 msg="system memory" total="62.8 GiB" free="57.5 GiB" free_swap="0 B"
  203. time=2025-03-01T19:55:30.800Z level=WARN source=ggml.go:132 msg="key not found" key=granite.attention.key_length default=64
  204. time=2025-03-01T19:55:30.800Z level=WARN source=ggml.go:132 msg="key not found" key=granite.attention.value_length default=64
  205. time=2025-03-01T19:55:30.801Z level=INFO source=server.go:130 msg=offload library=cuda layers.requested=-1 layers.model=41 layers.offload=41 layers.split="" memory.available="[22.2 GiB]" memory.gpu_overhead="0 B" memory.required.full="11.2 GiB" memory.required.partial="11.2 GiB" memory.required.kv="5.0 GiB" memory.required.allocations="[11.2 GiB]" memory.weights.total="6.4 GiB" memory.weights.repeating="6.3 GiB" memory.weights.nonrepeating="78.8 MiB" memory.graph.full="3.3 GiB" memory.graph.partial="3.3 GiB" projector.weights="851.2 MiB" projector.graph="0 B"
  206. time=2025-03-01T19:55:30.801Z level=INFO source=server.go:380 msg="starting llama server" cmd="/usr/bin/ollama runner --model /root/.ollama/models/blobs/sha256-1aefcd9a8a15091b670951963b5f8a7e6653bb1350345e9621e179685ac9bc5f --ctx-size 65536 --batch-size 512 --n-gpu-layers 41 --mmproj /root/.ollama/models/blobs/sha256-4d464be24899cf8dc1862945432e0cef4366c4181fa38b14754cc9279b727608 --threads 6 --parallel 4 --port 42981"
  207. time=2025-03-01T19:55:30.801Z level=INFO source=sched.go:450 msg="loaded runners" count=2
  208. time=2025-03-01T19:55:30.801Z level=INFO source=server.go:557 msg="waiting for llama runner to start responding"
  209. time=2025-03-01T19:55:30.802Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error"
  210. time=2025-03-01T19:55:30.833Z level=INFO source=runner.go:932 msg="starting go runner"
  211. ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
  212. ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
  213. ggml_cuda_init: found 1 CUDA devices:
  214. Device 0: NVIDIA GeForce RTX 4090, compute capability 8.9, VMM: yes
  215. load_backend: loaded CUDA backend from /usr/lib/ollama/cuda_v12/libggml-cuda.so
  216. load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-sandybridge.so
  217. time=2025-03-01T19:55:30.892Z level=INFO source=runner.go:935 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | CUDA : ARCHS = 600,610,620,700,720,750,800,860,870,890,900 | USE_GRAPHS = 1 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | LLAMAFILE = 1 | cgo(gcc)" threads=6
  218. time=2025-03-01T19:55:30.893Z level=INFO source=runner.go:993 msg="Server listening on 127.0.0.1:42981"
  219. llama_load_model_from_file: using device CUDA0 (NVIDIA GeForce RTX 4090) - 22761 MiB free
  220. llama_model_loader: loaded meta data with 32 key-value pairs and 362 tensors from /root/.ollama/models/blobs/sha256-1aefcd9a8a15091b670951963b5f8a7e6653bb1350345e9621e179685ac9bc5f (version GGUF V3 (latest))
  221. llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
  222. llama_model_loader: - kv 0: general.architecture str = granite
  223. llama_model_loader: - kv 1: general.type str = model
  224. llama_model_loader: - kv 2: general.name str = Granite_Vision_Llm
  225. llama_model_loader: - kv 3: general.size_label str = 2.5B
  226. llama_model_loader: - kv 4: granite.block_count u32 = 40
  227. llama_model_loader: - kv 5: granite.context_length u32 = 16384
  228. llama_model_loader: - kv 6: granite.embedding_length u32 = 2048
  229. llama_model_loader: - kv 7: granite.feed_forward_length u32 = 8192
  230. llama_model_loader: - kv 8: granite.attention.head_count u32 = 32
  231. llama_model_loader: - kv 9: granite.attention.head_count_kv u32 = 8
  232. llama_model_loader: - kv 10: granite.rope.freq_base f32 = 300000.000000
  233. llama_model_loader: - kv 11: granite.attention.layer_norm_rms_epsilon f32 = 0.000010
  234. llama_model_loader: - kv 12: granite.vocab_size u32 = 49156
  235. llama_model_loader: - kv 13: granite.rope.dimension_count u32 = 64
  236. llama_model_loader: - kv 14: granite.attention.scale f32 = 0.015625
  237. llama_model_loader: - kv 15: granite.embedding_scale f32 = 12.000000
  238. llama_model_loader: - kv 16: granite.residual_scale f32 = 0.220000
  239. llama_model_loader: - kv 17: granite.logit_scale f32 = 8.000000
  240. llama_model_loader: - kv 18: tokenizer.ggml.model str = gpt2
  241. llama_model_loader: - kv 19: tokenizer.ggml.pre str = refact
  242. llama_model_loader: - kv 20: tokenizer.ggml.tokens arr[str,49156] = ["<|end_of_text|>", "<fim_prefix>", "...
  243. llama_model_loader: - kv 21: tokenizer.ggml.token_type arr[i32,49156] = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ...
  244. llama_model_loader: - kv 22: tokenizer.ggml.merges arr[str,48891] = ["Ġ Ġ", "ĠĠ ĠĠ", "ĠĠĠĠ ĠĠ...
  245. llama_model_loader: - kv 23: tokenizer.ggml.bos_token_id u32 = 0
  246. llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 0
  247. llama_model_loader: - kv 25: tokenizer.ggml.unknown_token_id u32 = 0
  248. llama_model_loader: - kv 26: tokenizer.ggml.padding_token_id u32 = 0
  249. llama_model_loader: - kv 27: tokenizer.ggml.add_bos_token bool = false
  250. llama_model_loader: - kv 28: tokenizer.chat_template str = {%- if tools %}\n {{- '<|start_of_r...
  251. llama_model_loader: - kv 29: tokenizer.ggml.add_space_prefix bool = false
  252. llama_model_loader: - kv 30: general.quantization_version u32 = 2
  253. llama_model_loader: - kv 31: general.file_type u32 = 15
  254. llama_model_loader: - type f32: 81 tensors
  255. llama_model_loader: - type q4_K: 240 tensors
  256. llama_model_loader: - type q6_K: 41 tensors
  257. time=2025-03-01T19:55:31.052Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model"
  258. llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
  259. llm_load_vocab: special tokens cache size = 23
  260. llm_load_vocab: token to piece cache size = 0.2826 MB
  261. llm_load_print_meta: format = GGUF V3 (latest)
  262. llm_load_print_meta: arch = granite
  263. llm_load_print_meta: vocab type = BPE
  264. llm_load_print_meta: n_vocab = 49156
  265. llm_load_print_meta: n_merges = 48891
  266. llm_load_print_meta: vocab_only = 0
  267. llm_load_print_meta: n_ctx_train = 16384
  268. llm_load_print_meta: n_embd = 2048
  269. llm_load_print_meta: n_layer = 40
  270. llm_load_print_meta: n_head = 32
  271. llm_load_print_meta: n_head_kv = 8
  272. llm_load_print_meta: n_rot = 64
  273. llm_load_print_meta: n_swa = 0
  274. llm_load_print_meta: n_embd_head_k = 64
  275. llm_load_print_meta: n_embd_head_v = 64
  276. llm_load_print_meta: n_gqa = 4
  277. llm_load_print_meta: n_embd_k_gqa = 512
  278. llm_load_print_meta: n_embd_v_gqa = 512
  279. llm_load_print_meta: f_norm_eps = 0.0e+00
  280. llm_load_print_meta: f_norm_rms_eps = 1.0e-05
  281. llm_load_print_meta: f_clamp_kqv = 0.0e+00
  282. llm_load_print_meta: f_max_alibi_bias = 0.0e+00
  283. llm_load_print_meta: f_logit_scale = 8.0e+00
  284. llm_load_print_meta: n_ff = 8192
  285. llm_load_print_meta: n_expert = 0
  286. llm_load_print_meta: n_expert_used = 0
  287. llm_load_print_meta: causal attn = 1
  288. llm_load_print_meta: pooling type = 0
  289. llm_load_print_meta: rope type = 0
  290. llm_load_print_meta: rope scaling = linear
  291. llm_load_print_meta: freq_base_train = 300000.0
  292. llm_load_print_meta: freq_scale_train = 1
  293. llm_load_print_meta: n_ctx_orig_yarn = 16384
  294. llm_load_print_meta: rope_finetuned = unknown
  295. llm_load_print_meta: ssm_d_conv = 0
  296. llm_load_print_meta: ssm_d_inner = 0
  297. llm_load_print_meta: ssm_d_state = 0
  298. llm_load_print_meta: ssm_dt_rank = 0
  299. llm_load_print_meta: ssm_dt_b_c_rms = 0
  300. llm_load_print_meta: model type = 3B
  301. llm_load_print_meta: model ftype = Q4_K - Medium
  302. llm_load_print_meta: model params = 2.53 B
  303. llm_load_print_meta: model size = 1.44 GiB (4.87 BPW)
  304. llm_load_print_meta: general.name = Granite_Vision_Llm
  305. llm_load_print_meta: BOS token = 0 '<|end_of_text|>'
  306. llm_load_print_meta: EOS token = 0 '<|end_of_text|>'
  307. llm_load_print_meta: UNK token = 0 '<|end_of_text|>'
  308. llm_load_print_meta: PAD token = 0 '<|end_of_text|>'
  309. llm_load_print_meta: LF token = 145 'Ä'
  310. llm_load_print_meta: EOG token = 0 '<|end_of_text|>'
  311. llm_load_print_meta: max token length = 512
  312. llm_load_print_meta: f_embedding_scale = 12.000000
  313. llm_load_print_meta: f_residual_scale = 0.220000
  314. llm_load_print_meta: f_attention_scale = 0.015625
  315. [GIN] 2025/03/01 - 19:55:31 | 200 | 34.542µs | 192.168.17.19 | GET "/"
  316. llm_load_tensors: offloading 40 repeating layers to GPU
  317. llm_load_tensors: offloading output layer to GPU
  318. llm_load_tensors: offloaded 41/41 layers to GPU
  319. llm_load_tensors: CUDA0 model buffer size = 1472.05 MiB
  320. llm_load_tensors: CPU_Mapped model buffer size = 78.76 MiB
  321. llama_new_context_with_model: n_seq_max = 4
  322. llama_new_context_with_model: n_ctx = 65536
  323. llama_new_context_with_model: n_ctx_per_seq = 16384
  324. llama_new_context_with_model: n_batch = 2048
  325. llama_new_context_with_model: n_ubatch = 512
  326. llama_new_context_with_model: flash_attn = 0
  327. llama_new_context_with_model: freq_base = 300000.0
  328. llama_new_context_with_model: freq_scale = 1
  329. llama_kv_cache_init: kv_size = 65536, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 40, can_shift = 1
  330. llama_kv_cache_init: CUDA0 KV buffer size = 5120.00 MiB
  331. llama_new_context_with_model: KV self size = 5120.00 MiB, K (f16): 2560.00 MiB, V (f16): 2560.00 MiB
  332. llama_new_context_with_model: CUDA_Host output buffer size = 0.78 MiB
  333. llama_new_context_with_model: CUDA0 compute buffer size = 4240.00 MiB
  334. llama_new_context_with_model: CUDA_Host compute buffer size = 132.01 MiB
  335. llama_new_context_with_model: graph nodes = 1368
  336. llama_new_context_with_model: graph splits = 2
  337. key clip.use_silu not found in file
  338. key clip.vision.image_crop_resolution not found in file
  339. clip_model_load: model name: siglip-model
  340. clip_model_load: description: image encoder for LLaVA
  341. clip_model_load: GGUF version: 3
  342. clip_model_load: alignment: 32
  343. clip_model_load: n_tensors: 451
  344. clip_model_load: n_kv: 22
  345. clip_model_load: ftype: f16
  346.  
  347. clip_model_load: loaded meta data with 22 key-value pairs and 451 tensors from /root/.ollama/models/blobs/sha256-4d464be24899cf8dc1862945432e0cef4366c4181fa38b14754cc9279b727608
  348. clip_model_load: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
  349. clip_model_load: - kv 0: general.architecture str = clip
  350. clip_model_load: - kv 1: clip.has_text_encoder bool = false
  351. ggml.c:2705: GGML_ASSERT(ggml_can_mul_mat(a, b)) failed
  352. clip_model_load: - kv 2: clip.has_vision_encoder bool = true
  353. clip_model_load: - kv 3: clip.has_llava_projector bool = true
  354. clip_model_load: - kv 4: general.file_type u32 = 1
  355. clip_model_load: - kv 5: general.name str = siglip-model
  356. clip_model_load: - kv 6: general.description str = image encoder for LLaVA
  357. clip_model_load: - kv 7: clip.projector_type str = mlp
  358. clip_model_load: - kv 8: clip.vision.image_size u32 = 384
  359. clip_model_load: - kv 9: clip.vision.patch_size u32 = 14
  360. clip_model_load: - kv 10: clip.vision.embedding_length u32 = 1152
  361. clip_model_load: - kv 11: clip.vision.feed_forward_length u32 = 4304
  362. clip_model_load: - kv 12: clip.vision.projection_dim u32 = 0
  363. clip_model_load: - kv 13: clip.vision.attention.head_count u32 = 16
  364. clip_model_load: - kv 14: clip.vision.attention.layer_norm_epsilon f32 = 0.000001
  365. clip_model_load: - kv 15: clip.vision.block_count u32 = 27
  366. clip_model_load: - kv 16: clip.vision.image_grid_pinpoints arr[i32,52] = [384, 768, 384, 1152, 384, 1536, 384,...
  367. clip_model_load: - kv 17: clip.vision.mm_patch_merge_type str = spatial_unpad
  368. clip_model_load: - kv 18: clip.vision.feature_layer arr[i32,4] = [4, 8, 16, 27]
  369. clip_model_load: - kv 19: clip.vision.image_mean arr[f32,3] = [0.500000, 0.500000, 0.500000]
  370. clip_model_load: - kv 20: clip.vision.image_std arr[f32,3] = [0.500000, 0.500000, 0.500000]
  371. clip_model_load: - kv 21: clip.use_gelu bool = false
  372. clip_model_load: - type f32: 282 tensors
  373. clip_model_load: - type f16: 169 tensors
  374. clip_model_load: using CUDA0 backend
  375. clip_model_load: text_encoder: 0
  376. clip_model_load: vision_encoder: 1
  377. clip_model_load: llava_projector: 1
  378. clip_model_load: minicpmv_projector: 0
  379. clip_model_load: model size: 851.17 MB
  380. clip_model_load: metadata size: 0.16 MB
  381. clip_model_load: params backend buffer size = 851.17 MB (451 tensors)
  382. SIGSEGV: segmentation violation
  383. PC=0x7f743a858c47 m=4 sigcode=1 addr=0x206a03fe0
  384. signal arrived during cgo execution
  385.  
  386. goroutine 50 gp=0xc0005048c0 m=4 mp=0xc000079508 [syscall]:
  387. runtime.cgocall(0x565457c3c980, 0xc00008bc88)
  388. runtime/cgocall.go:167 +0x4b fp=0xc00008bc60 sp=0xc00008bc28 pc=0x565457027acb
  389. github.com/ollama/ollama/llama._Cfunc_clip_model_load(0x7f74589fbb70, 0x1)
  390. _cgo_gotypes.go:303 +0x50 fp=0xc00008bc88 sp=0xc00008bc60 pc=0x5654573dc830
  391. github.com/ollama/ollama/llama.NewClipContext(0xc0005206e0, {0x7ffe2e10b6b8, 0x62})
  392. github.com/ollama/ollama/llama/llama.go:512 +0x90 fp=0xc00008bd48 sp=0xc00008bc88 pc=0x5654573e3a90
  393. github.com/ollama/ollama/runner/llamarunner.NewImageContext(0xc0005206e0, {0x7ffe2e10b6b8, 0x62})
  394. github.com/ollama/ollama/runner/llamarunner/image.go:37 +0xf8 fp=0xc00008bdc8 sp=0xc00008bd48 pc=0x5654573f8ab8
  395. github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc00071a000, {0x29, 0x0, 0x1, 0x0, {0x0, 0x0, 0x0}, 0xc0005023a0, 0x0}, ...)
  396. github.com/ollama/ollama/runner/llamarunner/runner.go:871 +0x228 fp=0xc00008bf10 sp=0xc00008bdc8 pc=0x5654573fe4c8
  397. github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1()
  398. github.com/ollama/ollama/runner/llamarunner/runner.go:968 +0xda fp=0xc00008bfe0 sp=0xc00008bf10 pc=0x5654573ffc5a
  399. runtime.goexit({})
  400. runtime/asm_amd64.s:1700 +0x1 fp=0xc00008bfe8 sp=0xc00008bfe0 pc=0x5654570365a1
  401. created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
  402. github.com/ollama/ollama/runner/llamarunner/runner.go:968 +0xcd5
  403.  
  404. goroutine 1 gp=0xc0000061c0 m=nil [IO wait]:
  405. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  406. runtime/proc.go:424 +0xce fp=0xc0004b55c0 sp=0xc0004b55a0 pc=0x56545702e1ce
  407. runtime.netpollblock(0x20011bf80?, 0x56fc4fe6?, 0x54?)
  408. runtime/netpoll.go:575 +0xf7 fp=0xc0004b55f8 sp=0xc0004b55c0 pc=0x565456ff1e37
  409. internal/poll.runtime_pollWait(0x7f7461744680, 0x72)
  410. runtime/netpoll.go:351 +0x85 fp=0xc0004b5618 sp=0xc0004b55f8 pc=0x56545702d4c5
  411. internal/poll.(*pollDesc).wait(0xc000482080?, 0x2c?, 0x0)
  412. internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0004b5640 sp=0xc0004b5618 pc=0x5654570b5707
  413. internal/poll.(*pollDesc).waitRead(...)
  414. internal/poll/fd_poll_runtime.go:89
  415. internal/poll.(*FD).Accept(0xc000482080)
  416. internal/poll/fd_unix.go:620 +0x295 fp=0xc0004b56e8 sp=0xc0004b5640 pc=0x5654570baad5
  417. net.(*netFD).accept(0xc000482080)
  418. net/fd_unix.go:172 +0x29 fp=0xc0004b57a0 sp=0xc0004b56e8 pc=0x565457123bc9
  419. net.(*TCPListener).accept(0xc0000a0f40)
  420. net/tcpsock_posix.go:159 +0x1e fp=0xc0004b57f0 sp=0xc0004b57a0 pc=0x56545713983e
  421. net.(*TCPListener).Accept(0xc0000a0f40)
  422. net/tcpsock.go:372 +0x30 fp=0xc0004b5820 sp=0xc0004b57f0 pc=0x5654571386f0
  423. net/http.(*onceCloseListener).Accept(0xc00011a000?)
  424. <autogenerated>:1 +0x24 fp=0xc0004b5838 sp=0xc0004b5820 pc=0x565457382964
  425. net/http.(*Server).Serve(0xc0007163c0, {0x56545828d4f8, 0xc0000a0f40})
  426. net/http/server.go:3330 +0x30c fp=0xc0004b5968 sp=0xc0004b5838 pc=0x56545735a8ec
  427. github.com/ollama/ollama/runner/llamarunner.Execute({0xc000036140, 0x10, 0x10})
  428. github.com/ollama/ollama/runner/llamarunner/runner.go:994 +0x1174 fp=0xc0004b5d08 sp=0xc0004b5968 pc=0x5654573ff834
  429. github.com/ollama/ollama/runner.Execute({0xc000036130?, 0x0?, 0x0?})
  430. github.com/ollama/ollama/runner/runner.go:22 +0xd4 fp=0xc0004b5d30 sp=0xc0004b5d08 pc=0x56545762fc54
  431. github.com/ollama/ollama/cmd.NewCLI.func2(0xc0001dd200?, {0x565457e2a050?, 0x4?, 0x565457e2a054?})
  432. github.com/ollama/ollama/cmd/cmd.go:1280 +0x45 fp=0xc0004b5d58 sp=0xc0004b5d30 pc=0x565457c3c245
  433. github.com/spf13/cobra.(*Command).execute(0xc0001ebb08, {0xc0001dd400, 0x10, 0x10})
  434. github.com/spf13/[email protected]/command.go:940 +0x862 fp=0xc0004b5e78 sp=0xc0004b5d58 pc=0x56545719c902
  435. github.com/spf13/cobra.(*Command).ExecuteC(0xc00053f508)
  436. github.com/spf13/[email protected]/command.go:1068 +0x3a5 fp=0xc0004b5f30 sp=0xc0004b5e78 pc=0x56545719d145
  437. github.com/spf13/cobra.(*Command).Execute(...)
  438. github.com/spf13/[email protected]/command.go:992
  439. github.com/spf13/cobra.(*Command).ExecuteContext(...)
  440. github.com/spf13/[email protected]/command.go:985
  441. main.main()
  442. github.com/ollama/ollama/main.go:12 +0x4d fp=0xc0004b5f50 sp=0xc0004b5f30 pc=0x565457c3c5cd
  443. runtime.main()
  444. runtime/proc.go:272 +0x29d fp=0xc0004b5fe0 sp=0xc0004b5f50 pc=0x565456ff94dd
  445. runtime.goexit({})
  446. runtime/asm_amd64.s:1700 +0x1 fp=0xc0004b5fe8 sp=0xc0004b5fe0 pc=0x5654570365a1
  447.  
  448. goroutine 2 gp=0xc000006c40 m=nil [force gc (idle)]:
  449. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  450. runtime/proc.go:424 +0xce fp=0xc000072fa8 sp=0xc000072f88 pc=0x56545702e1ce
  451. runtime.goparkunlock(...)
  452. runtime/proc.go:430
  453. runtime.forcegchelper()
  454. runtime/proc.go:337 +0xb8 fp=0xc000072fe0 sp=0xc000072fa8 pc=0x565456ff9818
  455. runtime.goexit({})
  456. runtime/asm_amd64.s:1700 +0x1 fp=0xc000072fe8 sp=0xc000072fe0 pc=0x5654570365a1
  457. created by runtime.init.7 in goroutine 1
  458. runtime/proc.go:325 +0x1a
  459.  
  460. goroutine 3 gp=0xc000007180 m=nil [GC sweep wait]:
  461. runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
  462. runtime/proc.go:424 +0xce fp=0xc000073780 sp=0xc000073760 pc=0x56545702e1ce
  463. runtime.goparkunlock(...)
  464. runtime/proc.go:430
  465. runtime.bgsweep(0xc00003c080)
  466. runtime/mgcsweep.go:317 +0xdf fp=0xc0000737c8 sp=0xc000073780 pc=0x565456fe3ebf
  467. runtime.gcenable.gowrap1()
  468. runtime/mgc.go:204 +0x25 fp=0xc0000737e0 sp=0xc0000737c8 pc=0x565456fd8505
  469. runtime.goexit({})
  470. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000737e8 sp=0xc0000737e0 pc=0x5654570365a1
  471. created by runtime.gcenable in goroutine 1
  472. runtime/mgc.go:204 +0x66
  473.  
  474. goroutine 4 gp=0xc000007340 m=nil [GC scavenge wait]:
  475. runtime.gopark(0x10000?, 0x565457fdd6f8?, 0x0?, 0x0?, 0x0?)
  476. runtime/proc.go:424 +0xce fp=0xc000073f78 sp=0xc000073f58 pc=0x56545702e1ce
  477. runtime.goparkunlock(...)
  478. runtime/proc.go:430
  479. runtime.(*scavengerState).park(0x565458a84080)
  480. runtime/mgcscavenge.go:425 +0x49 fp=0xc000073fa8 sp=0xc000073f78 pc=0x565456fe1889
  481. runtime.bgscavenge(0xc00003c080)
  482. runtime/mgcscavenge.go:658 +0x59 fp=0xc000073fc8 sp=0xc000073fa8 pc=0x565456fe1e19
  483. runtime.gcenable.gowrap2()
  484. runtime/mgc.go:205 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x565456fd84a5
  485. runtime.goexit({})
  486. runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x5654570365a1
  487. created by runtime.gcenable in goroutine 1
  488. runtime/mgc.go:205 +0xa5
  489.  
  490. goroutine 5 gp=0xc000007c00 m=nil [finalizer wait]:
  491. runtime.gopark(0xc000072648?, 0x565456fcea05?, 0xb0?, 0x1?, 0xc0000061c0?)
  492. runtime/proc.go:424 +0xce fp=0xc000072620 sp=0xc000072600 pc=0x56545702e1ce
  493. runtime.runfinq()
  494. runtime/mfinal.go:193 +0x107 fp=0xc0000727e0 sp=0xc000072620 pc=0x565456fd7587
  495. runtime.goexit({})
  496. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000727e8 sp=0xc0000727e0 pc=0x5654570365a1
  497. created by runtime.createfing in goroutine 1
  498. runtime/mfinal.go:163 +0x3d
  499.  
  500. goroutine 6 gp=0xc0001cb500 m=nil [chan receive]:
  501. runtime.gopark(0xc000074760?, 0x56545710b245?, 0x60?, 0x69?, 0x5654582a2280?)
  502. runtime/proc.go:424 +0xce fp=0xc000074718 sp=0xc0000746f8 pc=0x56545702e1ce
  503. runtime.chanrecv(0xc000038380, 0x0, 0x1)
  504. runtime/chan.go:639 +0x41c fp=0xc000074790 sp=0xc000074718 pc=0x565456fc7bfc
  505. runtime.chanrecv1(0x0?, 0x0?)
  506. runtime/chan.go:489 +0x12 fp=0xc0000747b8 sp=0xc000074790 pc=0x565456fc77b2
  507. runtime.unique_runtime_registerUniqueMapCleanup.func1(...)
  508. runtime/mgc.go:1781
  509. runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
  510. runtime/mgc.go:1784 +0x2f fp=0xc0000747e0 sp=0xc0000747b8 pc=0x565456fdb56f
  511. runtime.goexit({})
  512. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000747e8 sp=0xc0000747e0 pc=0x5654570365a1
  513. created by unique.runtime_registerUniqueMapCleanup in goroutine 1
  514. runtime/mgc.go:1779 +0x96
  515.  
  516. goroutine 7 gp=0xc0001cba40 m=nil [GC worker (idle)]:
  517. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  518. runtime/proc.go:424 +0xce fp=0xc000074f38 sp=0xc000074f18 pc=0x56545702e1ce
  519. runtime.gcBgMarkWorker(0xc000039b20)
  520. runtime/mgc.go:1412 +0xe9 fp=0xc000074fc8 sp=0xc000074f38 pc=0x565456fda869
  521. runtime.gcBgMarkStartWorkers.gowrap1()
  522. runtime/mgc.go:1328 +0x25 fp=0xc000074fe0 sp=0xc000074fc8 pc=0x565456fda745
  523. runtime.goexit({})
  524. runtime/asm_amd64.s:1700 +0x1 fp=0xc000074fe8 sp=0xc000074fe0 pc=0x5654570365a1
  525. created by runtime.gcBgMarkStartWorkers in goroutine 1
  526. runtime/mgc.go:1328 +0x105
  527.  
  528. goroutine 18 gp=0xc000104380 m=nil [GC worker (idle)]:
  529. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  530. runtime/proc.go:424 +0xce fp=0xc00006e738 sp=0xc00006e718 pc=0x56545702e1ce
  531. runtime.gcBgMarkWorker(0xc000039b20)
  532. runtime/mgc.go:1412 +0xe9 fp=0xc00006e7c8 sp=0xc00006e738 pc=0x565456fda869
  533. runtime.gcBgMarkStartWorkers.gowrap1()
  534. runtime/mgc.go:1328 +0x25 fp=0xc00006e7e0 sp=0xc00006e7c8 pc=0x565456fda745
  535. runtime.goexit({})
  536. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006e7e8 sp=0xc00006e7e0 pc=0x5654570365a1
  537. created by runtime.gcBgMarkStartWorkers in goroutine 1
  538. runtime/mgc.go:1328 +0x105
  539.  
  540. goroutine 34 gp=0xc000504000 m=nil [GC worker (idle)]:
  541. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  542. runtime/proc.go:424 +0xce fp=0xc00050a738 sp=0xc00050a718 pc=0x56545702e1ce
  543. runtime.gcBgMarkWorker(0xc000039b20)
  544. runtime/mgc.go:1412 +0xe9 fp=0xc00050a7c8 sp=0xc00050a738 pc=0x565456fda869
  545. runtime.gcBgMarkStartWorkers.gowrap1()
  546. runtime/mgc.go:1328 +0x25 fp=0xc00050a7e0 sp=0xc00050a7c8 pc=0x565456fda745
  547. runtime.goexit({})
  548. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050a7e8 sp=0xc00050a7e0 pc=0x5654570365a1
  549. created by runtime.gcBgMarkStartWorkers in goroutine 1
  550. runtime/mgc.go:1328 +0x105
  551.  
  552. goroutine 8 gp=0xc0001cbc00 m=nil [GC worker (idle)]:
  553. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  554. runtime/proc.go:424 +0xce fp=0xc000075738 sp=0xc000075718 pc=0x56545702e1ce
  555. runtime.gcBgMarkWorker(0xc000039b20)
  556. runtime/mgc.go:1412 +0xe9 fp=0xc0000757c8 sp=0xc000075738 pc=0x565456fda869
  557. runtime.gcBgMarkStartWorkers.gowrap1()
  558. runtime/mgc.go:1328 +0x25 fp=0xc0000757e0 sp=0xc0000757c8 pc=0x565456fda745
  559. runtime.goexit({})
  560. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000757e8 sp=0xc0000757e0 pc=0x5654570365a1
  561. created by runtime.gcBgMarkStartWorkers in goroutine 1
  562. runtime/mgc.go:1328 +0x105
  563.  
  564. goroutine 19 gp=0xc000104540 m=nil [GC worker (idle)]:
  565. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  566. runtime/proc.go:424 +0xce fp=0xc00006ef38 sp=0xc00006ef18 pc=0x56545702e1ce
  567. runtime.gcBgMarkWorker(0xc000039b20)
  568. runtime/mgc.go:1412 +0xe9 fp=0xc00006efc8 sp=0xc00006ef38 pc=0x565456fda869
  569. runtime.gcBgMarkStartWorkers.gowrap1()
  570. runtime/mgc.go:1328 +0x25 fp=0xc00006efe0 sp=0xc00006efc8 pc=0x565456fda745
  571. runtime.goexit({})
  572. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006efe8 sp=0xc00006efe0 pc=0x5654570365a1
  573. created by runtime.gcBgMarkStartWorkers in goroutine 1
  574. runtime/mgc.go:1328 +0x105
  575.  
  576. goroutine 35 gp=0xc0005041c0 m=nil [GC worker (idle)]:
  577. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  578. runtime/proc.go:424 +0xce fp=0xc00050af38 sp=0xc00050af18 pc=0x56545702e1ce
  579. runtime.gcBgMarkWorker(0xc000039b20)
  580. runtime/mgc.go:1412 +0xe9 fp=0xc00050afc8 sp=0xc00050af38 pc=0x565456fda869
  581. runtime.gcBgMarkStartWorkers.gowrap1()
  582. runtime/mgc.go:1328 +0x25 fp=0xc00050afe0 sp=0xc00050afc8 pc=0x565456fda745
  583. runtime.goexit({})
  584. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050afe8 sp=0xc00050afe0 pc=0x5654570365a1
  585. created by runtime.gcBgMarkStartWorkers in goroutine 1
  586. runtime/mgc.go:1328 +0x105
  587.  
  588. goroutine 9 gp=0xc0001cbdc0 m=nil [GC worker (idle)]:
  589. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  590. runtime/proc.go:424 +0xce fp=0xc000075f38 sp=0xc000075f18 pc=0x56545702e1ce
  591. runtime.gcBgMarkWorker(0xc000039b20)
  592. runtime/mgc.go:1412 +0xe9 fp=0xc000075fc8 sp=0xc000075f38 pc=0x565456fda869
  593. runtime.gcBgMarkStartWorkers.gowrap1()
  594. runtime/mgc.go:1328 +0x25 fp=0xc000075fe0 sp=0xc000075fc8 pc=0x565456fda745
  595. runtime.goexit({})
  596. runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x5654570365a1
  597. created by runtime.gcBgMarkStartWorkers in goroutine 1
  598. runtime/mgc.go:1328 +0x105
  599.  
  600. goroutine 20 gp=0xc000104700 m=nil [GC worker (idle)]:
  601. runtime.gopark(0x565458b326e0?, 0x1?, 0xe0?, 0x68?, 0x0?)
  602. runtime/proc.go:424 +0xce fp=0xc00006f738 sp=0xc00006f718 pc=0x56545702e1ce
  603. runtime.gcBgMarkWorker(0xc000039b20)
  604. runtime/mgc.go:1412 +0xe9 fp=0xc00006f7c8 sp=0xc00006f738 pc=0x565456fda869
  605. runtime.gcBgMarkStartWorkers.gowrap1()
  606. runtime/mgc.go:1328 +0x25 fp=0xc00006f7e0 sp=0xc00006f7c8 pc=0x565456fda745
  607. runtime.goexit({})
  608. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006f7e8 sp=0xc00006f7e0 pc=0x5654570365a1
  609. created by runtime.gcBgMarkStartWorkers in goroutine 1
  610. runtime/mgc.go:1328 +0x105
  611.  
  612. goroutine 36 gp=0xc000504380 m=nil [GC worker (idle)]:
  613. runtime.gopark(0x5d4f68f97487a?, 0x0?, 0x0?, 0x0?, 0x0?)
  614. runtime/proc.go:424 +0xce fp=0xc00050b738 sp=0xc00050b718 pc=0x56545702e1ce
  615. runtime.gcBgMarkWorker(0xc000039b20)
  616. runtime/mgc.go:1412 +0xe9 fp=0xc00050b7c8 sp=0xc00050b738 pc=0x565456fda869
  617. runtime.gcBgMarkStartWorkers.gowrap1()
  618. runtime/mgc.go:1328 +0x25 fp=0xc00050b7e0 sp=0xc00050b7c8 pc=0x565456fda745
  619. runtime.goexit({})
  620. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050b7e8 sp=0xc00050b7e0 pc=0x5654570365a1
  621. created by runtime.gcBgMarkStartWorkers in goroutine 1
  622. runtime/mgc.go:1328 +0x105
  623.  
  624. goroutine 10 gp=0xc0004a4000 m=nil [GC worker (idle)]:
  625. runtime.gopark(0x565458b326e0?, 0x1?, 0xbd?, 0x5c?, 0x0?)
  626. runtime/proc.go:424 +0xce fp=0xc000506738 sp=0xc000506718 pc=0x56545702e1ce
  627. runtime.gcBgMarkWorker(0xc000039b20)
  628. runtime/mgc.go:1412 +0xe9 fp=0xc0005067c8 sp=0xc000506738 pc=0x565456fda869
  629. runtime.gcBgMarkStartWorkers.gowrap1()
  630. runtime/mgc.go:1328 +0x25 fp=0xc0005067e0 sp=0xc0005067c8 pc=0x565456fda745
  631. runtime.goexit({})
  632. runtime/asm_amd64.s:1700 +0x1 fp=0xc0005067e8 sp=0xc0005067e0 pc=0x5654570365a1
  633. created by runtime.gcBgMarkStartWorkers in goroutine 1
  634. runtime/mgc.go:1328 +0x105
  635.  
  636. goroutine 21 gp=0xc0001048c0 m=nil [GC worker (idle)]:
  637. runtime.gopark(0x5d4f68f973b5f?, 0x0?, 0x0?, 0x0?, 0x0?)
  638. runtime/proc.go:424 +0xce fp=0xc00006ff38 sp=0xc00006ff18 pc=0x56545702e1ce
  639. runtime.gcBgMarkWorker(0xc000039b20)
  640. runtime/mgc.go:1412 +0xe9 fp=0xc00006ffc8 sp=0xc00006ff38 pc=0x565456fda869
  641. runtime.gcBgMarkStartWorkers.gowrap1()
  642. runtime/mgc.go:1328 +0x25 fp=0xc00006ffe0 sp=0xc00006ffc8 pc=0x565456fda745
  643. runtime.goexit({})
  644. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006ffe8 sp=0xc00006ffe0 pc=0x5654570365a1
  645. created by runtime.gcBgMarkStartWorkers in goroutine 1
  646. runtime/mgc.go:1328 +0x105
  647.  
  648. goroutine 37 gp=0xc000504540 m=nil [GC worker (idle)]:
  649. runtime.gopark(0x5d4f68f97c38d?, 0x3?, 0xb3?, 0x20?, 0x0?)
  650. runtime/proc.go:424 +0xce fp=0xc00050bf38 sp=0xc00050bf18 pc=0x56545702e1ce
  651. runtime.gcBgMarkWorker(0xc000039b20)
  652. runtime/mgc.go:1412 +0xe9 fp=0xc00050bfc8 sp=0xc00050bf38 pc=0x565456fda869
  653. runtime.gcBgMarkStartWorkers.gowrap1()
  654. runtime/mgc.go:1328 +0x25 fp=0xc00050bfe0 sp=0xc00050bfc8 pc=0x565456fda745
  655. runtime.goexit({})
  656. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050bfe8 sp=0xc00050bfe0 pc=0x5654570365a1
  657. created by runtime.gcBgMarkStartWorkers in goroutine 1
  658. runtime/mgc.go:1328 +0x105
  659.  
  660. goroutine 51 gp=0xc000504a80 m=nil [semacquire]:
  661. runtime.gopark(0x0?, 0x0?, 0x60?, 0x0?, 0x0?)
  662. runtime/proc.go:424 +0xce fp=0xc000506e18 sp=0xc000506df8 pc=0x56545702e1ce
  663. runtime.goparkunlock(...)
  664. runtime/proc.go:430
  665. runtime.semacquire1(0xc00071a008, 0x0, 0x1, 0x0, 0x12)
  666. runtime/sema.go:178 +0x22c fp=0xc000506e80 sp=0xc000506e18 pc=0x56545700c58c
  667. sync.runtime_Semacquire(0x0?)
  668. runtime/sema.go:71 +0x25 fp=0xc000506eb8 sp=0xc000506e80 pc=0x56545702f9e5
  669. sync.(*WaitGroup).Wait(0x0?)
  670. sync/waitgroup.go:118 +0x48 fp=0xc000506ee0 sp=0xc000506eb8 pc=0x565457044ee8
  671. github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc00071a000, {0x56545828f920, 0xc0000e6b40})
  672. github.com/ollama/ollama/runner/llamarunner/runner.go:316 +0x47 fp=0xc000506fb8 sp=0xc000506ee0 pc=0x5654573fab27
  673. github.com/ollama/ollama/runner/llamarunner.Execute.gowrap2()
  674. github.com/ollama/ollama/runner/llamarunner/runner.go:973 +0x28 fp=0xc000506fe0 sp=0xc000506fb8 pc=0x5654573ffb48
  675. runtime.goexit({})
  676. runtime/asm_amd64.s:1700 +0x1 fp=0xc000506fe8 sp=0xc000506fe0 pc=0x5654570365a1
  677. created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
  678. github.com/ollama/ollama/runner/llamarunner/runner.go:973 +0xdb5
  679.  
  680. goroutine 38 gp=0xc000584380 m=nil [IO wait]:
  681. runtime.gopark(0x5654570b8d05?, 0xc0001b8000?, 0x10?, 0x9a?, 0xb?)
  682. runtime/proc.go:424 +0xce fp=0xc000139918 sp=0xc0001398f8 pc=0x56545702e1ce
  683. runtime.netpollblock(0x5654570516b8?, 0x56fc4fe6?, 0x54?)
  684. runtime/netpoll.go:575 +0xf7 fp=0xc000139950 sp=0xc000139918 pc=0x565456ff1e37
  685. internal/poll.runtime_pollWait(0x7f7461744568, 0x72)
  686. runtime/netpoll.go:351 +0x85 fp=0xc000139970 sp=0xc000139950 pc=0x56545702d4c5
  687. internal/poll.(*pollDesc).wait(0xc0001b8000?, 0xc000130000?, 0x0)
  688. internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc000139998 sp=0xc000139970 pc=0x5654570b5707
  689. internal/poll.(*pollDesc).waitRead(...)
  690. internal/poll/fd_poll_runtime.go:89
  691. internal/poll.(*FD).Read(0xc0001b8000, {0xc000130000, 0x1000, 0x1000})
  692. internal/poll/fd_unix.go:165 +0x27a fp=0xc000139a30 sp=0xc000139998 pc=0x5654570b69fa
  693. net.(*netFD).Read(0xc0001b8000, {0xc000130000?, 0xc000139aa0?, 0x5654570b5bc5?})
  694. net/fd_posix.go:55 +0x25 fp=0xc000139a78 sp=0xc000139a30 pc=0x565457121c05
  695. net.(*conn).Read(0xc000118000, {0xc000130000?, 0x0?, 0xc0001160c8?})
  696. net/net.go:189 +0x45 fp=0xc000139ac0 sp=0xc000139a78 pc=0x565457130205
  697. net.(*TCPConn).Read(0xc0001160c0?, {0xc000130000?, 0xc0001b8000?, 0xc000139af8?})
  698. <autogenerated>:1 +0x25 fp=0xc000139af0 sp=0xc000139ac0 pc=0x565457143405
  699. net/http.(*connReader).Read(0xc0001160c0, {0xc000130000, 0x1000, 0x1000})
  700. net/http/server.go:798 +0x14b fp=0xc000139b40 sp=0xc000139af0 pc=0x5654573506ab
  701. bufio.(*Reader).fill(0xc000536060)
  702. bufio/bufio.go:110 +0x103 fp=0xc000139b78 sp=0xc000139b40 pc=0x565457147b03
  703. bufio.(*Reader).Peek(0xc000536060, 0x4)
  704. bufio/bufio.go:148 +0x53 fp=0xc000139b98 sp=0xc000139b78 pc=0x565457147c33
  705. net/http.(*conn).serve(0xc00011a000, {0x56545828f8e8, 0xc000718300})
  706. net/http/server.go:2127 +0x738 fp=0xc000139fb8 sp=0xc000139b98 pc=0x5654573559f8
  707. net/http.(*Server).Serve.gowrap3()
  708. net/http/server.go:3360 +0x28 fp=0xc000139fe0 sp=0xc000139fb8 pc=0x56545735ace8
  709. runtime.goexit({})
  710. runtime/asm_amd64.s:1700 +0x1 fp=0xc000139fe8 sp=0xc000139fe0 pc=0x5654570365a1
  711. created by net/http.(*Server).Serve in goroutine 1
  712. net/http/server.go:3360 +0x485
  713.  
  714. rax 0x206a03fe0
  715. rbx 0x7f7458170580
  716. rcx 0xff8
  717. rdx 0x7f7458008980
  718. rdi 0x7f7458008990
  719. rsi 0x0
  720. rbp 0x7f7460f3b5b0
  721. rsp 0x7f7460f3b590
  722. r8 0x0
  723. r9 0x7f74500d1a28
  724. r10 0x0
  725. r11 0x246
  726. r12 0x7f7459e8f380
  727. r13 0x7f7458008990
  728. r14 0x0
  729. r15 0x565491c52f70
  730. rip 0x7f743a858c47
  731. rflags 0x10297
  732. cs 0x33
  733. fs 0x0
  734. gs 0x0
  735. SIGABRT: abort
  736. PC=0x7f74a8a5700b m=4 sigcode=18446744073709551610
  737. signal arrived during cgo execution
  738.  
  739. goroutine 50 gp=0xc0005048c0 m=4 mp=0xc000079508 [syscall]:
  740. runtime.cgocall(0x565457c3c980, 0xc00008bc88)
  741. runtime/cgocall.go:167 +0x4b fp=0xc00008bc60 sp=0xc00008bc28 pc=0x565457027acb
  742. github.com/ollama/ollama/llama._Cfunc_clip_model_load(0x7f74589fbb70, 0x1)
  743. _cgo_gotypes.go:303 +0x50 fp=0xc00008bc88 sp=0xc00008bc60 pc=0x5654573dc830
  744. github.com/ollama/ollama/llama.NewClipContext(0xc0005206e0, {0x7ffe2e10b6b8, 0x62})
  745. github.com/ollama/ollama/llama/llama.go:512 +0x90 fp=0xc00008bd48 sp=0xc00008bc88 pc=0x5654573e3a90
  746. github.com/ollama/ollama/runner/llamarunner.NewImageContext(0xc0005206e0, {0x7ffe2e10b6b8, 0x62})
  747. github.com/ollama/ollama/runner/llamarunner/image.go:37 +0xf8 fp=0xc00008bdc8 sp=0xc00008bd48 pc=0x5654573f8ab8
  748. github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc00071a000, {0x29, 0x0, 0x1, 0x0, {0x0, 0x0, 0x0}, 0xc0005023a0, 0x0}, ...)
  749. github.com/ollama/ollama/runner/llamarunner/runner.go:871 +0x228 fp=0xc00008bf10 sp=0xc00008bdc8 pc=0x5654573fe4c8
  750. github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1()
  751. github.com/ollama/ollama/runner/llamarunner/runner.go:968 +0xda fp=0xc00008bfe0 sp=0xc00008bf10 pc=0x5654573ffc5a
  752. runtime.goexit({})
  753. runtime/asm_amd64.s:1700 +0x1 fp=0xc00008bfe8 sp=0xc00008bfe0 pc=0x5654570365a1
  754. created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
  755. github.com/ollama/ollama/runner/llamarunner/runner.go:968 +0xcd5
  756.  
  757. goroutine 1 gp=0xc0000061c0 m=nil [IO wait]:
  758. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  759. runtime/proc.go:424 +0xce fp=0xc0004b55c0 sp=0xc0004b55a0 pc=0x56545702e1ce
  760. runtime.netpollblock(0x20011bf80?, 0x56fc4fe6?, 0x54?)
  761. runtime/netpoll.go:575 +0xf7 fp=0xc0004b55f8 sp=0xc0004b55c0 pc=0x565456ff1e37
  762. internal/poll.runtime_pollWait(0x7f7461744680, 0x72)
  763. runtime/netpoll.go:351 +0x85 fp=0xc0004b5618 sp=0xc0004b55f8 pc=0x56545702d4c5
  764. internal/poll.(*pollDesc).wait(0xc000482080?, 0x2c?, 0x0)
  765. internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0004b5640 sp=0xc0004b5618 pc=0x5654570b5707
  766. internal/poll.(*pollDesc).waitRead(...)
  767. internal/poll/fd_poll_runtime.go:89
  768. internal/poll.(*FD).Accept(0xc000482080)
  769. internal/poll/fd_unix.go:620 +0x295 fp=0xc0004b56e8 sp=0xc0004b5640 pc=0x5654570baad5
  770. net.(*netFD).accept(0xc000482080)
  771. net/fd_unix.go:172 +0x29 fp=0xc0004b57a0 sp=0xc0004b56e8 pc=0x565457123bc9
  772. net.(*TCPListener).accept(0xc0000a0f40)
  773. net/tcpsock_posix.go:159 +0x1e fp=0xc0004b57f0 sp=0xc0004b57a0 pc=0x56545713983e
  774. net.(*TCPListener).Accept(0xc0000a0f40)
  775. net/tcpsock.go:372 +0x30 fp=0xc0004b5820 sp=0xc0004b57f0 pc=0x5654571386f0
  776. net/http.(*onceCloseListener).Accept(0xc00011a000?)
  777. <autogenerated>:1 +0x24 fp=0xc0004b5838 sp=0xc0004b5820 pc=0x565457382964
  778. net/http.(*Server).Serve(0xc0007163c0, {0x56545828d4f8, 0xc0000a0f40})
  779. net/http/server.go:3330 +0x30c fp=0xc0004b5968 sp=0xc0004b5838 pc=0x56545735a8ec
  780. github.com/ollama/ollama/runner/llamarunner.Execute({0xc000036140, 0x10, 0x10})
  781. github.com/ollama/ollama/runner/llamarunner/runner.go:994 +0x1174 fp=0xc0004b5d08 sp=0xc0004b5968 pc=0x5654573ff834
  782. github.com/ollama/ollama/runner.Execute({0xc000036130?, 0x0?, 0x0?})
  783. github.com/ollama/ollama/runner/runner.go:22 +0xd4 fp=0xc0004b5d30 sp=0xc0004b5d08 pc=0x56545762fc54
  784. github.com/ollama/ollama/cmd.NewCLI.func2(0xc0001dd200?, {0x565457e2a050?, 0x4?, 0x565457e2a054?})
  785. github.com/ollama/ollama/cmd/cmd.go:1280 +0x45 fp=0xc0004b5d58 sp=0xc0004b5d30 pc=0x565457c3c245
  786. github.com/spf13/cobra.(*Command).execute(0xc0001ebb08, {0xc0001dd400, 0x10, 0x10})
  787. github.com/spf13/[email protected]/command.go:940 +0x862 fp=0xc0004b5e78 sp=0xc0004b5d58 pc=0x56545719c902
  788. github.com/spf13/cobra.(*Command).ExecuteC(0xc00053f508)
  789. github.com/spf13/[email protected]/command.go:1068 +0x3a5 fp=0xc0004b5f30 sp=0xc0004b5e78 pc=0x56545719d145
  790. github.com/spf13/cobra.(*Command).Execute(...)
  791. github.com/spf13/[email protected]/command.go:992
  792. github.com/spf13/cobra.(*Command).ExecuteContext(...)
  793. github.com/spf13/[email protected]/command.go:985
  794. main.main()
  795. github.com/ollama/ollama/main.go:12 +0x4d fp=0xc0004b5f50 sp=0xc0004b5f30 pc=0x565457c3c5cd
  796. runtime.main()
  797. runtime/proc.go:272 +0x29d fp=0xc0004b5fe0 sp=0xc0004b5f50 pc=0x565456ff94dd
  798. runtime.goexit({})
  799. runtime/asm_amd64.s:1700 +0x1 fp=0xc0004b5fe8 sp=0xc0004b5fe0 pc=0x5654570365a1
  800.  
  801. goroutine 2 gp=0xc000006c40 m=nil [force gc (idle)]:
  802. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  803. runtime/proc.go:424 +0xce fp=0xc000072fa8 sp=0xc000072f88 pc=0x56545702e1ce
  804. runtime.goparkunlock(...)
  805. runtime/proc.go:430
  806. runtime.forcegchelper()
  807. runtime/proc.go:337 +0xb8 fp=0xc000072fe0 sp=0xc000072fa8 pc=0x565456ff9818
  808. runtime.goexit({})
  809. runtime/asm_amd64.s:1700 +0x1 fp=0xc000072fe8 sp=0xc000072fe0 pc=0x5654570365a1
  810. created by runtime.init.7 in goroutine 1
  811. runtime/proc.go:325 +0x1a
  812.  
  813. goroutine 3 gp=0xc000007180 m=nil [GC sweep wait]:
  814. runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
  815. runtime/proc.go:424 +0xce fp=0xc000073780 sp=0xc000073760 pc=0x56545702e1ce
  816. runtime.goparkunlock(...)
  817. runtime/proc.go:430
  818. runtime.bgsweep(0xc00003c080)
  819. runtime/mgcsweep.go:317 +0xdf fp=0xc0000737c8 sp=0xc000073780 pc=0x565456fe3ebf
  820. runtime.gcenable.gowrap1()
  821. runtime/mgc.go:204 +0x25 fp=0xc0000737e0 sp=0xc0000737c8 pc=0x565456fd8505
  822. runtime.goexit({})
  823. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000737e8 sp=0xc0000737e0 pc=0x5654570365a1
  824. created by runtime.gcenable in goroutine 1
  825. runtime/mgc.go:204 +0x66
  826.  
  827. goroutine 4 gp=0xc000007340 m=nil [GC scavenge wait]:
  828. runtime.gopark(0x10000?, 0x565457fdd6f8?, 0x0?, 0x0?, 0x0?)
  829. runtime/proc.go:424 +0xce fp=0xc000073f78 sp=0xc000073f58 pc=0x56545702e1ce
  830. runtime.goparkunlock(...)
  831. runtime/proc.go:430
  832. runtime.(*scavengerState).park(0x565458a84080)
  833. runtime/mgcscavenge.go:425 +0x49 fp=0xc000073fa8 sp=0xc000073f78 pc=0x565456fe1889
  834. runtime.bgscavenge(0xc00003c080)
  835. runtime/mgcscavenge.go:658 +0x59 fp=0xc000073fc8 sp=0xc000073fa8 pc=0x565456fe1e19
  836. runtime.gcenable.gowrap2()
  837. runtime/mgc.go:205 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x565456fd84a5
  838. runtime.goexit({})
  839. runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x5654570365a1
  840. created by runtime.gcenable in goroutine 1
  841. runtime/mgc.go:205 +0xa5
  842.  
  843. goroutine 5 gp=0xc000007c00 m=nil [finalizer wait]:
  844. runtime.gopark(0xc000072648?, 0x565456fcea05?, 0xb0?, 0x1?, 0xc0000061c0?)
  845. runtime/proc.go:424 +0xce fp=0xc000072620 sp=0xc000072600 pc=0x56545702e1ce
  846. runtime.runfinq()
  847. runtime/mfinal.go:193 +0x107 fp=0xc0000727e0 sp=0xc000072620 pc=0x565456fd7587
  848. runtime.goexit({})
  849. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000727e8 sp=0xc0000727e0 pc=0x5654570365a1
  850. created by runtime.createfing in goroutine 1
  851. runtime/mfinal.go:163 +0x3d
  852.  
  853. goroutine 6 gp=0xc0001cb500 m=nil [chan receive]:
  854. runtime.gopark(0xc000074760?, 0x56545710b245?, 0x60?, 0x69?, 0x5654582a2280?)
  855. runtime/proc.go:424 +0xce fp=0xc000074718 sp=0xc0000746f8 pc=0x56545702e1ce
  856. runtime.chanrecv(0xc000038380, 0x0, 0x1)
  857. runtime/chan.go:639 +0x41c fp=0xc000074790 sp=0xc000074718 pc=0x565456fc7bfc
  858. runtime.chanrecv1(0x0?, 0x0?)
  859. runtime/chan.go:489 +0x12 fp=0xc0000747b8 sp=0xc000074790 pc=0x565456fc77b2
  860. runtime.unique_runtime_registerUniqueMapCleanup.func1(...)
  861. runtime/mgc.go:1781
  862. runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
  863. runtime/mgc.go:1784 +0x2f fp=0xc0000747e0 sp=0xc0000747b8 pc=0x565456fdb56f
  864. runtime.goexit({})
  865. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000747e8 sp=0xc0000747e0 pc=0x5654570365a1
  866. created by unique.runtime_registerUniqueMapCleanup in goroutine 1
  867. runtime/mgc.go:1779 +0x96
  868.  
  869. goroutine 7 gp=0xc0001cba40 m=nil [GC worker (idle)]:
  870. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  871. runtime/proc.go:424 +0xce fp=0xc000074f38 sp=0xc000074f18 pc=0x56545702e1ce
  872. runtime.gcBgMarkWorker(0xc000039b20)
  873. runtime/mgc.go:1412 +0xe9 fp=0xc000074fc8 sp=0xc000074f38 pc=0x565456fda869
  874. runtime.gcBgMarkStartWorkers.gowrap1()
  875. runtime/mgc.go:1328 +0x25 fp=0xc000074fe0 sp=0xc000074fc8 pc=0x565456fda745
  876. runtime.goexit({})
  877. runtime/asm_amd64.s:1700 +0x1 fp=0xc000074fe8 sp=0xc000074fe0 pc=0x5654570365a1
  878. created by runtime.gcBgMarkStartWorkers in goroutine 1
  879. runtime/mgc.go:1328 +0x105
  880.  
  881. goroutine 18 gp=0xc000104380 m=nil [GC worker (idle)]:
  882. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  883. runtime/proc.go:424 +0xce fp=0xc00006e738 sp=0xc00006e718 pc=0x56545702e1ce
  884. runtime.gcBgMarkWorker(0xc000039b20)
  885. runtime/mgc.go:1412 +0xe9 fp=0xc00006e7c8 sp=0xc00006e738 pc=0x565456fda869
  886. runtime.gcBgMarkStartWorkers.gowrap1()
  887. runtime/mgc.go:1328 +0x25 fp=0xc00006e7e0 sp=0xc00006e7c8 pc=0x565456fda745
  888. runtime.goexit({})
  889. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006e7e8 sp=0xc00006e7e0 pc=0x5654570365a1
  890. created by runtime.gcBgMarkStartWorkers in goroutine 1
  891. runtime/mgc.go:1328 +0x105
  892.  
  893. goroutine 34 gp=0xc000504000 m=nil [GC worker (idle)]:
  894. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  895. runtime/proc.go:424 +0xce fp=0xc00050a738 sp=0xc00050a718 pc=0x56545702e1ce
  896. runtime.gcBgMarkWorker(0xc000039b20)
  897. runtime/mgc.go:1412 +0xe9 fp=0xc00050a7c8 sp=0xc00050a738 pc=0x565456fda869
  898. runtime.gcBgMarkStartWorkers.gowrap1()
  899. runtime/mgc.go:1328 +0x25 fp=0xc00050a7e0 sp=0xc00050a7c8 pc=0x565456fda745
  900. runtime.goexit({})
  901. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050a7e8 sp=0xc00050a7e0 pc=0x5654570365a1
  902. created by runtime.gcBgMarkStartWorkers in goroutine 1
  903. runtime/mgc.go:1328 +0x105
  904.  
  905. goroutine 8 gp=0xc0001cbc00 m=nil [GC worker (idle)]:
  906. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  907. runtime/proc.go:424 +0xce fp=0xc000075738 sp=0xc000075718 pc=0x56545702e1ce
  908. runtime.gcBgMarkWorker(0xc000039b20)
  909. runtime/mgc.go:1412 +0xe9 fp=0xc0000757c8 sp=0xc000075738 pc=0x565456fda869
  910. runtime.gcBgMarkStartWorkers.gowrap1()
  911. runtime/mgc.go:1328 +0x25 fp=0xc0000757e0 sp=0xc0000757c8 pc=0x565456fda745
  912. runtime.goexit({})
  913. runtime/asm_amd64.s:1700 +0x1 fp=0xc0000757e8 sp=0xc0000757e0 pc=0x5654570365a1
  914. created by runtime.gcBgMarkStartWorkers in goroutine 1
  915. runtime/mgc.go:1328 +0x105
  916.  
  917. goroutine 19 gp=0xc000104540 m=nil [GC worker (idle)]:
  918. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  919. runtime/proc.go:424 +0xce fp=0xc00006ef38 sp=0xc00006ef18 pc=0x56545702e1ce
  920. runtime.gcBgMarkWorker(0xc000039b20)
  921. runtime/mgc.go:1412 +0xe9 fp=0xc00006efc8 sp=0xc00006ef38 pc=0x565456fda869
  922. runtime.gcBgMarkStartWorkers.gowrap1()
  923. runtime/mgc.go:1328 +0x25 fp=0xc00006efe0 sp=0xc00006efc8 pc=0x565456fda745
  924. runtime.goexit({})
  925. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006efe8 sp=0xc00006efe0 pc=0x5654570365a1
  926. created by runtime.gcBgMarkStartWorkers in goroutine 1
  927. runtime/mgc.go:1328 +0x105
  928.  
  929. goroutine 35 gp=0xc0005041c0 m=nil [GC worker (idle)]:
  930. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  931. runtime/proc.go:424 +0xce fp=0xc00050af38 sp=0xc00050af18 pc=0x56545702e1ce
  932. runtime.gcBgMarkWorker(0xc000039b20)
  933. runtime/mgc.go:1412 +0xe9 fp=0xc00050afc8 sp=0xc00050af38 pc=0x565456fda869
  934. runtime.gcBgMarkStartWorkers.gowrap1()
  935. runtime/mgc.go:1328 +0x25 fp=0xc00050afe0 sp=0xc00050afc8 pc=0x565456fda745
  936. runtime.goexit({})
  937. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050afe8 sp=0xc00050afe0 pc=0x5654570365a1
  938. created by runtime.gcBgMarkStartWorkers in goroutine 1
  939. runtime/mgc.go:1328 +0x105
  940.  
  941. goroutine 9 gp=0xc0001cbdc0 m=nil [GC worker (idle)]:
  942. runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
  943. runtime/proc.go:424 +0xce fp=0xc000075f38 sp=0xc000075f18 pc=0x56545702e1ce
  944. runtime.gcBgMarkWorker(0xc000039b20)
  945. runtime/mgc.go:1412 +0xe9 fp=0xc000075fc8 sp=0xc000075f38 pc=0x565456fda869
  946. runtime.gcBgMarkStartWorkers.gowrap1()
  947. runtime/mgc.go:1328 +0x25 fp=0xc000075fe0 sp=0xc000075fc8 pc=0x565456fda745
  948. runtime.goexit({})
  949. runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x5654570365a1
  950. created by runtime.gcBgMarkStartWorkers in goroutine 1
  951. runtime/mgc.go:1328 +0x105
  952.  
  953. goroutine 20 gp=0xc000104700 m=nil [GC worker (idle)]:
  954. runtime.gopark(0x565458b326e0?, 0x1?, 0xe0?, 0x68?, 0x0?)
  955. runtime/proc.go:424 +0xce fp=0xc00006f738 sp=0xc00006f718 pc=0x56545702e1ce
  956. runtime.gcBgMarkWorker(0xc000039b20)
  957. runtime/mgc.go:1412 +0xe9 fp=0xc00006f7c8 sp=0xc00006f738 pc=0x565456fda869
  958. runtime.gcBgMarkStartWorkers.gowrap1()
  959. runtime/mgc.go:1328 +0x25 fp=0xc00006f7e0 sp=0xc00006f7c8 pc=0x565456fda745
  960. runtime.goexit({})
  961. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006f7e8 sp=0xc00006f7e0 pc=0x5654570365a1
  962. created by runtime.gcBgMarkStartWorkers in goroutine 1
  963. runtime/mgc.go:1328 +0x105
  964.  
  965. goroutine 36 gp=0xc000504380 m=nil [GC worker (idle)]:
  966. runtime.gopark(0x5d4f68f97487a?, 0x0?, 0x0?, 0x0?, 0x0?)
  967. runtime/proc.go:424 +0xce fp=0xc00050b738 sp=0xc00050b718 pc=0x56545702e1ce
  968. runtime.gcBgMarkWorker(0xc000039b20)
  969. runtime/mgc.go:1412 +0xe9 fp=0xc00050b7c8 sp=0xc00050b738 pc=0x565456fda869
  970. runtime.gcBgMarkStartWorkers.gowrap1()
  971. runtime/mgc.go:1328 +0x25 fp=0xc00050b7e0 sp=0xc00050b7c8 pc=0x565456fda745
  972. runtime.goexit({})
  973. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050b7e8 sp=0xc00050b7e0 pc=0x5654570365a1
  974. created by runtime.gcBgMarkStartWorkers in goroutine 1
  975. runtime/mgc.go:1328 +0x105
  976.  
  977. goroutine 10 gp=0xc0004a4000 m=nil [GC worker (idle)]:
  978. runtime.gopark(0x565458b326e0?, 0x1?, 0xbd?, 0x5c?, 0x0?)
  979. runtime/proc.go:424 +0xce fp=0xc000506738 sp=0xc000506718 pc=0x56545702e1ce
  980. runtime.gcBgMarkWorker(0xc000039b20)
  981. runtime/mgc.go:1412 +0xe9 fp=0xc0005067c8 sp=0xc000506738 pc=0x565456fda869
  982. runtime.gcBgMarkStartWorkers.gowrap1()
  983. runtime/mgc.go:1328 +0x25 fp=0xc0005067e0 sp=0xc0005067c8 pc=0x565456fda745
  984. runtime.goexit({})
  985. runtime/asm_amd64.s:1700 +0x1 fp=0xc0005067e8 sp=0xc0005067e0 pc=0x5654570365a1
  986. created by runtime.gcBgMarkStartWorkers in goroutine 1
  987. runtime/mgc.go:1328 +0x105
  988.  
  989. goroutine 21 gp=0xc0001048c0 m=nil [GC worker (idle)]:
  990. runtime.gopark(0x5d4f68f973b5f?, 0x0?, 0x0?, 0x0?, 0x0?)
  991. runtime/proc.go:424 +0xce fp=0xc00006ff38 sp=0xc00006ff18 pc=0x56545702e1ce
  992. runtime.gcBgMarkWorker(0xc000039b20)
  993. runtime/mgc.go:1412 +0xe9 fp=0xc00006ffc8 sp=0xc00006ff38 pc=0x565456fda869
  994. runtime.gcBgMarkStartWorkers.gowrap1()
  995. runtime/mgc.go:1328 +0x25 fp=0xc00006ffe0 sp=0xc00006ffc8 pc=0x565456fda745
  996. runtime.goexit({})
  997. runtime/asm_amd64.s:1700 +0x1 fp=0xc00006ffe8 sp=0xc00006ffe0 pc=0x5654570365a1
  998. created by runtime.gcBgMarkStartWorkers in goroutine 1
  999. runtime/mgc.go:1328 +0x105
  1000.  
  1001. goroutine 37 gp=0xc000504540 m=nil [GC worker (idle)]:
  1002. runtime.gopark(0x5d4f68f97c38d?, 0x3?, 0xb3?, 0x20?, 0x0?)
  1003. runtime/proc.go:424 +0xce fp=0xc00050bf38 sp=0xc00050bf18 pc=0x56545702e1ce
  1004. runtime.gcBgMarkWorker(0xc000039b20)
  1005. runtime/mgc.go:1412 +0xe9 fp=0xc00050bfc8 sp=0xc00050bf38 pc=0x565456fda869
  1006. runtime.gcBgMarkStartWorkers.gowrap1()
  1007. runtime/mgc.go:1328 +0x25 fp=0xc00050bfe0 sp=0xc00050bfc8 pc=0x565456fda745
  1008. runtime.goexit({})
  1009. runtime/asm_amd64.s:1700 +0x1 fp=0xc00050bfe8 sp=0xc00050bfe0 pc=0x5654570365a1
  1010. created by runtime.gcBgMarkStartWorkers in goroutine 1
  1011. runtime/mgc.go:1328 +0x105
  1012.  
  1013. goroutine 51 gp=0xc000504a80 m=nil [semacquire]:
  1014. runtime.gopark(0x0?, 0x0?, 0x60?, 0x0?, 0x0?)
  1015. runtime/proc.go:424 +0xce fp=0xc000506e18 sp=0xc000506df8 pc=0x56545702e1ce
  1016. runtime.goparkunlock(...)
  1017. runtime/proc.go:430
  1018. runtime.semacquire1(0xc00071a008, 0x0, 0x1, 0x0, 0x12)
  1019. runtime/sema.go:178 +0x22c fp=0xc000506e80 sp=0xc000506e18 pc=0x56545700c58c
  1020. sync.runtime_Semacquire(0x0?)
  1021. runtime/sema.go:71 +0x25 fp=0xc000506eb8 sp=0xc000506e80 pc=0x56545702f9e5
  1022. sync.(*WaitGroup).Wait(0x0?)
  1023. sync/waitgroup.go:118 +0x48 fp=0xc000506ee0 sp=0xc000506eb8 pc=0x565457044ee8
  1024. github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc00071a000, {0x56545828f920, 0xc0000e6b40})
  1025. github.com/ollama/ollama/runner/llamarunner/runner.go:316 +0x47 fp=0xc000506fb8 sp=0xc000506ee0 pc=0x5654573fab27
  1026. github.com/ollama/ollama/runner/llamarunner.Execute.gowrap2()
  1027. github.com/ollama/ollama/runner/llamarunner/runner.go:973 +0x28 fp=0xc000506fe0 sp=0xc000506fb8 pc=0x5654573ffb48
  1028. runtime.goexit({})
  1029. runtime/asm_amd64.s:1700 +0x1 fp=0xc000506fe8 sp=0xc000506fe0 pc=0x5654570365a1
  1030. created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
  1031. github.com/ollama/ollama/runner/llamarunner/runner.go:973 +0xdb5
  1032.  
  1033. goroutine 38 gp=0xc000584380 m=nil [IO wait]:
  1034. runtime.gopark(0x5654570b8d05?, 0xc0001b8000?, 0x10?, 0x9a?, 0xb?)
  1035. runtime/proc.go:424 +0xce fp=0xc000139918 sp=0xc0001398f8 pc=0x56545702e1ce
  1036. runtime.netpollblock(0x5654570516b8?, 0x56fc4fe6?, 0x54?)
  1037. runtime/netpoll.go:575 +0xf7 fp=0xc000139950 sp=0xc000139918 pc=0x565456ff1e37
  1038. internal/poll.runtime_pollWait(0x7f7461744568, 0x72)
  1039. runtime/netpoll.go:351 +0x85 fp=0xc000139970 sp=0xc000139950 pc=0x56545702d4c5
  1040. internal/poll.(*pollDesc).wait(0xc0001b8000?, 0xc000130000?, 0x0)
  1041. internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc000139998 sp=0xc000139970 pc=0x5654570b5707
  1042. internal/poll.(*pollDesc).waitRead(...)
  1043. internal/poll/fd_poll_runtime.go:89
  1044. internal/poll.(*FD).Read(0xc0001b8000, {0xc000130000, 0x1000, 0x1000})
  1045. internal/poll/fd_unix.go:165 +0x27a fp=0xc000139a30 sp=0xc000139998 pc=0x5654570b69fa
  1046. net.(*netFD).Read(0xc0001b8000, {0xc000130000?, 0xc000139aa0?, 0x5654570b5bc5?})
  1047. net/fd_posix.go:55 +0x25 fp=0xc000139a78 sp=0xc000139a30 pc=0x565457121c05
  1048. net.(*conn).Read(0xc000118000, {0xc000130000?, 0x0?, 0xc0001160c8?})
  1049. net/net.go:189 +0x45 fp=0xc000139ac0 sp=0xc000139a78 pc=0x565457130205
  1050. net.(*TCPConn).Read(0xc0001160c0?, {0xc000130000?, 0xc0001b8000?, 0xc000139af8?})
  1051. <autogenerated>:1 +0x25 fp=0xc000139af0 sp=0xc000139ac0 pc=0x565457143405
  1052. net/http.(*connReader).Read(0xc0001160c0, {0xc000130000, 0x1000, 0x1000})
  1053. net/http/server.go:798 +0x14b fp=0xc000139b40 sp=0xc000139af0 pc=0x5654573506ab
  1054. bufio.(*Reader).fill(0xc000536060)
  1055. bufio/bufio.go:110 +0x103 fp=0xc000139b78 sp=0xc000139b40 pc=0x565457147b03
  1056. bufio.(*Reader).Peek(0xc000536060, 0x4)
  1057. bufio/bufio.go:148 +0x53 fp=0xc000139b98 sp=0xc000139b78 pc=0x565457147c33
  1058. net/http.(*conn).serve(0xc00011a000, {0x56545828f8e8, 0xc000718300})
  1059. net/http/server.go:2127 +0x738 fp=0xc000139fb8 sp=0xc000139b98 pc=0x5654573559f8
  1060. net/http.(*Server).Serve.gowrap3()
  1061. net/http/server.go:3360 +0x28 fp=0xc000139fe0 sp=0xc000139fb8 pc=0x56545735ace8
  1062. runtime.goexit({})
  1063. runtime/asm_amd64.s:1700 +0x1 fp=0xc000139fe8 sp=0xc000139fe0 pc=0x5654570365a1
  1064. created by net/http.(*Server).Serve in goroutine 1
  1065. net/http/server.go:3360 +0x485
  1066.  
  1067. rax 0x0
  1068. rbx 0x7f7460f3c700
  1069. rcx 0x7f74a8a5700b
  1070. rdx 0x0
  1071. rdi 0x2
  1072. rsi 0x7f7460f3b5c0
  1073. rbp 0x565457ffa96d
  1074. rsp 0x7f7460f3b5c0
  1075. r8 0x0
  1076. r9 0x7f7460f3b5c0
  1077. r10 0x8
  1078. r11 0x246
  1079. r12 0x56545802b14e
  1080. r13 0xa91
  1081. r14 0x7f745a157ce0
  1082. r15 0x48
  1083. rip 0x7f74a8a5700b
  1084. rflags 0x246
  1085. cs 0x33
  1086. fs 0x0
  1087. gs 0x0
  1088. time=2025-03-01T19:55:32.800Z level=ERROR source=server.go:421 msg="llama runner terminated" error="exit status 2"
  1089. time=2025-03-01T19:55:32.808Z level=ERROR source=sched.go:456 msg="error loading llama server" error="llama runner process has terminated: GGML_ASSERT(ggml_can_mul_mat(a, b)) failed"
  1090. [GIN] 2025/03/01 - 19:55:32 | 500 | 2.386495341s | 10.244.243.205 | POST "/api/chat"
  1091. [GIN] 2025/03/01 - 19:55:36 | 200 | 30.417µs | 192.168.17.19 | GET "/"
  1092. [GIN] 2025/03/01 - 19:55:36 | 200 | 31.334µs | 192.168.17.19 | GET "/"
  1093. time=2025-03-01T19:55:38.006Z level=WARN source=sched.go:647 msg="gpu VRAM usage didn't recover within timeout" seconds=5.197506217 model=/root/.ollama/models/blobs/sha256-1aefcd9a8a15091b670951963b5f8a7e6653bb1350345e9621e179685ac9bc5f
  1094. time=2025-03-01T19:55:38.255Z level=WARN source=sched.go:647 msg="gpu VRAM usage didn't recover within timeout" seconds=5.446975626 model=/root/.ollama/models/blobs/sha256-1aefcd9a8a15091b670951963b5f8a7e6653bb1350345e9621e179685ac9bc5f
  1095. time=2025-03-01T19:55:38.506Z level=WARN source=sched.go:647 msg="gpu VRAM usage didn't recover within timeout" seconds=5.697440302 model=/root/.ollama/models/blobs/sha256-1aefcd9a8a15091b670951963b5f8a7e6653bb1350345e9621e179685ac9bc5f
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!