github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11418] Llama Runner process has terminated [MSTY] #69596

New Issue
Closed
opened 2026-05-04 18:36:05 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-05-04 18:36:05 -05:00
Owner
Copy Link

Originally created by @psi00 on GitHub (Jul 14, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11418

What is the issue?

I'm using Msty.app as the UI, which uses Ollama for its backend. Running any model results in a Llama Runner error;
llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed

Relevant log output

{"level":50,"time":1752516964568,"pid":10104,"hostname":"Ryzen7","msg":"2025/07/14 19:16:04 routes.go:1232: 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_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:10000 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:3 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:C:\\\\Users\\\\James\\\\AppData\\\\Roaming\\\\Msty\\\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:3 OLLAMA_ORIGINS:[http://localhost http://127.0.0.1 http://0.0.0.0 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://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]\"\n"}
{"level":50,"time":1752516964570,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.569+01:00 level=INFO source=images.go:458 msg=\"total blobs: 10\"\n"}
{"level":50,"time":1752516964570,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.570+01:00 level=INFO source=images.go:465 msg=\"total unused blobs removed: 0\"\n"}
{"level":30,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"[GIN-debug] [WARNING] Creating an Engine instance with the Logger and Recovery middleware already attached.\n\n[GIN-debug] [WARNING] Running in \"debug\" mode. Switch to \"release\" mode in production.\n - using env:\texport GIN_MODE=release\n - using code:\tgin.SetMode(gin.ReleaseMode)\n\n"}
{"level":30,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"[GIN-debug] HEAD   /                         --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func1 (5 handlers)\n[GIN-debug] GET    /                         --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func2 (5 handlers)\n[GIN-debug] HEAD   /api/version              --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func3 (5 handlers)\n[GIN-debug] GET    /api/version              --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func4 (5 handlers)\n[GIN-debug] POST   /api/pull                 --> github.com/ollama/ollama/server.(*Server).PullHandler-fm (5 handlers)\n[GIN-debug] POST   /api/push                 --> github.com/ollama/ollama/server.(*Server).PushHandler-fm (5 handlers)\n[GIN-debug] HEAD   /api/tags                 --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers)\n[GIN-debug] GET    /api/tags                 --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers)\n[GIN-debug] POST   /api/show                 --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (5 handlers)\n[GIN-debug] DELETE /api/delete               --> github.com/ollama/ollama/server.(*Server).DeleteHandler-fm (5 handlers)\n[GIN-debug] POST   /api/create               --> github.com/ollama/ollama/server.(*Server).CreateHandler-fm (5 handlers)\n[GIN-debug] POST   /api/blobs/:digest        --> github.com/ollama/ollama/server.(*Server).CreateBlobHandler-fm (5 handlers)\n[GIN-debug] HEAD   /api/blobs/:digest        --> github.com/ollama/ollama/server.(*Server).HeadBlobHandler-fm (5 handlers)\n[GIN-debug] POST   /api/copy                 --> github.com/ollama/ollama/server.(*Server).CopyHandler-fm (5 handlers)\n[GIN-debug] GET    /api/ps                   --> github.com/ollama/ollama/server.(*Server).PsHandler-fm (5 handlers)\n[GIN-debug] POST   /api/generate             --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (5 handlers)\n[GIN-debug] POST   /api/chat                 --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (5 handlers)\n[GIN-debug] POST   /api/embed                --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (5 handlers)\n[GIN-debug] POST   /api/embeddings           --> github.com/ollama/ollama/server.(*Server).EmbeddingsHandler-fm (5 handlers)\n[GIN-debug] POST   /v1/chat/completions      --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (6 handlers)\n[GIN-debug] POST   /v1/completions           --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (6 handlers)\n[GIN-debug] POST   /v1/embeddings            --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (6 handlers)\n[GIN-debug] GET    /v1/models                --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (6 handlers)\n[GIN-debug] GET    /v1/models/:model         --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (6 handlers)\n"}
{"level":50,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.571+01:00 level=INFO source=routes.go:1299 msg=\"Listening on 127.0.0.1:10000 (version 0.6.6)\"\ntime=2025-07-14T19:16:04.571+01:00 level=INFO source=gpu.go:217 msg=\"looking for compatible GPUs\"\n"}
{"level":50,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.571+01:00 level=INFO source=gpu_windows.go:167 msg=packages count=1\ntime=2025-07-14T19:16:04.571+01:00 level=INFO source=gpu_windows.go:214 msg=\"\" package=0 cores=8 efficiency=0 threads=16\n"}
{"level":50,"time":1752516964691,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.691+01:00 level=INFO source=types.go:130 msg=\"inference compute\" id=GPU-8e0431b3-3727-b8f6-f64e-9303bcf98020 library=cuda variant=v12 compute=8.6 driver=12.8 name=\"NVIDIA GeForce RTX 3060\" total=\"12.0 GiB\" available=\"11.0 GiB\"\n"}
{"level":30,"time":1752516964798,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:04 | 200 |            0s |       127.0.0.1 | GET      \"/\"\n"}
{"level":30,"time":1752516965114,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 |      1.0282ms |       127.0.0.1 | GET      \"/api/tags\"\n"}
{"level":30,"time":1752516965590,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 |            0s |       127.0.0.1 | GET      \"/\"\n"}
{"level":30,"time":1752516965707,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 |            0s |       127.0.0.1 | GET      \"/\"\n"}
{"level":30,"time":1752516965742,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 |            0s |       127.0.0.1 | GET      \"/\"\n"}
{"level":30,"time":1752516965836,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 |            0s |       127.0.0.1 | GET      \"/\"\n"}
{"level":30,"time":1752516965839,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 |      1.5081ms |       127.0.0.1 | GET      \"/api/tags\"\n"}
{"level":50,"time":1752516971235,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.235+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=general.alignment default=32\n"}
{"level":50,"time":1752516971265,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.265+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=general.alignment default=32\n"}
{"level":50,"time":1752516971277,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.277+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=general.alignment default=32\n"}
{"level":50,"time":1752516971278,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.278+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.vision.block_count default=0\n"}
{"level":50,"time":1752516971301,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\ntime=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\n"}
{"level":50,"time":1752516971301,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.301+01:00 level=INFO source=sched.go:722 msg=\"new model will fit in available VRAM in single GPU, loading\" model=C:\\Users\\James\\AppData\\Roaming\\Msty\\models\\blobs\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc gpu=GPU-8e0431b3-3727-b8f6-f64e-9303bcf98020 parallel=3 available=11478908928 required=\"1.8 GiB\"\n"}
{"level":50,"time":1752516971316,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.315+01:00 level=INFO source=server.go:105 msg=\"system memory\" total=\"15.9 GiB\" free=\"5.0 GiB\" free_swap=\"14.3 GiB\"\ntime=2025-07-14T19:16:11.315+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.vision.block_count default=0\n"}
{"level":50,"time":1752516971331,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\n"}
{"level":50,"time":1752516971332,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.331+01:00 level=INFO source=server.go:138 msg=offload library=cuda layers.requested=-1 layers.model=29 layers.offload=29 layers.split=\"\" memory.available=\"[10.7 GiB]\" memory.gpu_overhead=\"0 B\" memory.required.full=\"1.8 GiB\" memory.required.partial=\"1.8 GiB\" memory.required.kv=\"168.0 MiB\" memory.required.allocations=\"[1.8 GiB]\" memory.weights.total=\"934.7 MiB\" memory.weights.repeating=\"752.1 MiB\" memory.weights.nonrepeating=\"182.6 MiB\" memory.graph.full=\"299.8 MiB\" memory.graph.partial=\"482.3 MiB\"\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=INFO source=server.go:185 msg=\"enabling flash attention\"\n"}
{"level":50,"time":1752516971332,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.331+01:00 level=WARN source=server.go:193 msg=\"kv cache type not supported by model\" type=\"\"\n"}
{"level":50,"time":1752516971368,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: loaded meta data with 26 key-value pairs and 339 tensors from C:\\Users\\James\\AppData\\Roaming\\Msty\\models\\blobs\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc (version GGUF V3 (latest))\nllama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.\n"}
{"level":50,"time":1752516971368,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv   0:                       general.architecture str              = qwen2\nllama_model_loader: - kv   1:                               general.type str              = model\nllama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 1.5B\nllama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen\nllama_model_loader: - kv   4:                         general.size_label str              = 1.5B\nllama_model_loader: - kv   5:                          qwen2.block_count u32              = 28\nllama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072\nllama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 1536\nllama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 8960\nllama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 12\nllama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 2\nllama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 10000.000000\nllama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000001\nllama_model_loader: - kv  13:                          general.file_type u32              = 15\nllama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2\nllama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = qwen2\n"}
{"level":50,"time":1752516971382,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,151936]  = [\"!\", \"\\\"\", \"#\", \"$\", \"%\", \"&\", \"'\", ...\n"}
{"level":50,"time":1752516971385,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...\n"}
{"level":50,"time":1752516971399,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = [\"Ġ Ġ\", \"ĠĠ ĠĠ\", \"i n\", \"Ġ t\",...\nllama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646\nllama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643\nllama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643\nllama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true\nllama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false\nllama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...\nllama_model_loader: - kv  25:               general.quantization_version u32              = 2\nllama_model_loader: - type  f32:  141 tensors\nllama_model_loader: - type q4_K:  169 tensors\nllama_model_loader: - type q6_K:   29 tensors\n"}
{"level":50,"time":1752516971400,"pid":10104,"hostname":"Ryzen7","msg":"print_info: file format = GGUF V3 (latest)\nprint_info: file type   = Q4_K - Medium\nprint_info: file size   = 1.04 GiB (5.00 BPW) \n"}
{"level":50,"time":1752516971496,"pid":10104,"hostname":"Ryzen7","msg":"load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect\n"}
{"level":50,"time":1752516971496,"pid":10104,"hostname":"Ryzen7","msg":"load: special tokens cache size = 22\n"}
{"level":50,"time":1752516971533,"pid":10104,"hostname":"Ryzen7","msg":"load: token to piece cache size = 0.9310 MB\nprint_info: arch             = qwen2\nprint_info: vocab_only       = 1\nprint_info: model type       = ?B\nprint_info: model params     = 1.78 B\nprint_info: general.name     = DeepSeek R1 Distill Qwen 1.5B\nprint_info: vocab type       = BPE\nprint_info: n_vocab          = 151936\nprint_info: n_merges         = 151387\nprint_info: BOS token        = 151646 '<｜begin▁of▁sentence｜>'\nprint_info: EOS token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOT token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: PAD token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: LF token         = 198 'Ċ'\nprint_info: FIM PRE token    = 151659 '<|fim_prefix|>'\nprint_info: FIM SUF token    = 151661 '<|fim_suffix|>'\n"}
{"level":50,"time":1752516971533,"pid":10104,"hostname":"Ryzen7","msg":"print_info: FIM MID token    = 151660 '<|fim_middle|>'\nprint_info: FIM PAD token    = 151662 '<|fim_pad|>'\nprint_info: FIM REP token    = 151663 '<|repo_name|>'\nprint_info: FIM SEP token    = 151664 '<|file_sep|>'\nprint_info: EOG token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOG token        = 151662 '<|fim_pad|>'\nprint_info: EOG token        = 151663 '<|repo_name|>'\nprint_info: EOG token        = 151664 '<|file_sep|>'\nprint_info: max token length = 256\nllama_model_load: vocab only - skipping tensors\n"}
{"level":50,"time":1752516971543,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.543+01:00 level=INFO source=server.go:405 msg=\"starting llama server\" cmd=\"C:\\\\Users\\\\James\\\\AppData\\\\Roaming\\\\Msty\\\\msty-local.exe runner --model C:\\\\Users\\\\James\\\\AppData\\\\Roaming\\\\Msty\\\\models\\\\blobs\\\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc --ctx-size 6144 --batch-size 512 --n-gpu-layers 29 --threads 8 --flash-attn --no-mmap --parallel 3 --port 59174\"\n"}
{"level":50,"time":1752516971548,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.547+01:00 level=INFO source=sched.go:451 msg=\"loaded runners\" count=1\ntime=2025-07-14T19:16:11.547+01:00 level=INFO source=server.go:580 msg=\"waiting for llama runner to start responding\"\n"}
{"level":50,"time":1752516971548,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.548+01:00 level=INFO source=server.go:614 msg=\"waiting for server to become available\" status=\"llm server error\"\n"}
{"level":50,"time":1752516971582,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.581+01:00 level=INFO source=runner.go:853 msg=\"starting go runner\"\n"}
{"level":50,"time":1752516971699,"pid":10104,"hostname":"Ryzen7","msg":"ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no\r\nggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no\r\nggml_cuda_init: found 1 CUDA devices:\r\n  Device 0: NVIDIA GeForce RTX 3060, compute capability 8.6, VMM: yes\r\nload_backend: loaded CUDA backend from C:\\Users\\James\\AppData\\Roaming\\Msty\\lib\\ollama\\cuda_v12\\ggml-cuda.dll\n"}
{"level":50,"time":1752516971707,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.706+01:00 level=INFO source=ggml.go:109 msg=system CPU.0.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)\n"}
{"level":50,"time":1752516971708,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.708+01:00 level=INFO source=runner.go:913 msg=\"Server listening on 127.0.0.1:59174\"\n"}
{"level":50,"time":1752516971789,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 3060) - 11242 MiB free\n"}
{"level":50,"time":1752516971799,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.798+01:00 level=INFO source=server.go:614 msg=\"waiting for server to become available\" status=\"llm server loading model\"\n"}
{"level":50,"time":1752516971826,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: loaded meta data with 26 key-value pairs and 339 tensors from C:\\Users\\James\\AppData\\Roaming\\Msty\\models\\blobs\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc (version GGUF V3 (latest))\nllama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.\nllama_model_loader: - kv   0:                       general.architecture str              = qwen2\nllama_model_loader: - kv   1:                               general.type str              = model\nllama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 1.5B\nllama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen\nllama_model_loader: - kv   4:                         general.size_label str              = 1.5B\nllama_model_loader: - kv   5:                          qwen2.block_count u32              = 28\nllama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072\nllama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 1536\nllama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 8960\nllama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 12\nllama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 2\nllama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 10000.000000\nllama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000001\nllama_model_loader: - kv  13:                          general.file_type u32              = 15\nllama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2\nllama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = qwen2\n"}
{"level":50,"time":1752516971840,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,151936]  = [\"!\", \"\\\"\", \"#\", \"$\", \"%\", \"&\", \"'\", ...\n"}
{"level":50,"time":1752516971844,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...\n"}
{"level":50,"time":1752516971859,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = [\"Ġ Ġ\", \"ĠĠ ĠĠ\", \"i n\", \"Ġ t\",...\nllama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646\nllama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643\nllama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643\nllama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true\nllama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false\nllama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...\nllama_model_loader: - kv  25:               general.quantization_version u32              = 2\nllama_model_loader: - type  f32:  141 tensors\nllama_model_loader: - type q4_K:  169 tensors\nllama_model_loader: - type q6_K:   29 tensors\nprint_info: file format = GGUF V3 (latest)\nprint_info: file type   = Q4_K - Medium\nprint_info: file size   = 1.04 GiB (5.00 BPW) \n"}
{"level":50,"time":1752516971940,"pid":10104,"hostname":"Ryzen7","msg":"load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect\n"}
{"level":50,"time":1752516971940,"pid":10104,"hostname":"Ryzen7","msg":"load: special tokens cache size = 22\n"}
{"level":50,"time":1752516971977,"pid":10104,"hostname":"Ryzen7","msg":"load: token to piece cache size = 0.9310 MB\nprint_info: arch             = qwen2\nprint_info: vocab_only       = 0\nprint_info: n_ctx_train      = 131072\nprint_info: n_embd           = 1536\nprint_info: n_layer          = 28\nprint_info: n_head           = 12\nprint_info: n_head_kv        = 2\nprint_info: n_rot            = 128\nprint_info: n_swa            = 0\nprint_info: n_swa_pattern    = 1\nprint_info: n_embd_head_k    = 128\nprint_info: n_embd_head_v    = 128\nprint_info: n_gqa            = 6\nprint_info: n_embd_k_gqa     = 256\nprint_info: n_embd_v_gqa     = 256\nprint_info: f_norm_eps       = 0.0e+00\nprint_info: f_norm_rms_eps   = 1.0e-06\nprint_info: f_clamp_kqv      = 0.0e+00\nprint_info: f_max_alibi_bias = 0.0e+00\nprint_info: f_logit_scale    = 0.0e+00\nprint_info: f_attn_scale     = 0.0e+00\nprint_info: n_ff             = 8960\nprint_info: n_expert         = 0\nprint_info: n_expert_used    = 0\nprint_info: causal attn      = 1\nprint_info: pooling type     = 0\nprint_info: rope type        = 2\nprint_info: rope scaling     = linear\nprint_info: freq_base_train  = 10000.0\nprint_info: freq_scale_train = 1\nprint_info: n_ctx_orig_yarn  = 131072\nprint_info: rope_finetuned   = unknown\nprint_info: ssm_d_conv       = 0\nprint_info: ssm_d_inner      = 0\nprint_info: ssm_d_state      = 0\nprint_info: ssm_dt_rank      = 0\nprint_info: ssm_dt_b_c_rms   = 0\nprint_info: model type       = 1.5B\nprint_info: model params     = 1.78 B\nprint_info: general.name     = DeepSeek R1 Distill Qwen 1.5B\nprint_info: vocab type       = BPE\nprint_info: n_vocab          = 151936\nprint_info: n_merges         = 151387\nprint_info: BOS token        = 151646 '<｜begin▁of▁sentence｜>'\nprint_info: EOS token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOT token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: PAD token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: LF token         = 198 'Ċ'\nprint_info: FIM PRE token    = 151659 '<|fim_prefix|>'\nprint_info: FIM SUF token    = 151661 '<|fim_suffix|>'\nprint_info: FIM MID token    = 151660 '<|fim_middle|>'\nprint_info: FIM PAD token    = 151662 '<|fim_pad|>'\nprint_info: FIM REP token    = 151663 '<|repo_name|>'\nprint_info: FIM SEP token    = 151664 '<|file_sep|>'\nprint_info: EOG token        = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOG token        = 151662 '<|fim_pad|>'\nprint_info: EOG token        = 151663 '<|repo_name|>'\nprint_info: EOG token        = 151664 '<|file_sep|>'\nprint_info: max token length = 256\nload_tensors: loading model tensors, this can take a while... (mmap = false)\n"}
{"level":50,"time":1752516971982,"pid":10104,"hostname":"Ryzen7","msg":"load_tensors: offloading 28 repeating layers to GPU\nload_tensors: offloading output layer to GPU\nload_tensors: offloaded 29/29 layers to GPU\nload_tensors:          CPU model buffer size =   125.19 MiB\nload_tensors:        CUDA0 model buffer size =   934.70 MiB\n"}
{"level":50,"time":1752516972202,"pid":10104,"hostname":"Ryzen7","msg":"llama_context: constructing llama_context\nllama_context: n_seq_max     = 3\nllama_context: n_ctx         = 6144\nllama_context: n_ctx_per_seq = 2048\nllama_context: n_batch       = 1536\nllama_context: n_ubatch      = 512\nllama_context: causal_attn   = 1\nllama_context: flash_attn    = 1\nllama_context: freq_base     = 10000.0\nllama_context: freq_scale    = 1\nllama_context: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized\n"}
{"level":50,"time":1752516972203,"pid":10104,"hostname":"Ryzen7","msg":"llama_context:  CUDA_Host  output buffer size =     1.76 MiB\ninit: kv_size = 6144, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1\n"}
{"level":50,"time":1752516972209,"pid":10104,"hostname":"Ryzen7","msg":"init:      CUDA0 KV buffer size =   168.00 MiB\nllama_context: KV self size  =  168.00 MiB, K (f16):   84.00 MiB, V (f16):   84.00 MiB\n"}
{"level":50,"time":1752516972217,"pid":10104,"hostname":"Ryzen7","msg":"D:/a/llama.cpp/llama.cpp/ggml/src/ggml.c:1777: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed\r\n"}
{"level":50,"time":1752516972501,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:12.500+01:00 level=INFO source=server.go:614 msg=\"waiting for server to become available\" status=\"llm server not responding\"\n"}
{"level":50,"time":1752516972565,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:12.565+01:00 level=ERROR source=server.go:449 msg=\"llama runner terminated\" error=\"exit status 0xc0000409\"\n"}
{"level":50,"time":1752516972751,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:12.750+01:00 level=ERROR source=sched.go:457 msg=\"error loading llama server\" error=\"llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed\"\n"}
{"level":30,"time":1752516972752,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:12 | 500 |    1.5308332s |       127.0.0.1 | POST     \"/api/chat\"\n"}

OS

Windows

GPU

AMD

CPU

AMD

Ollama version

0.6.6

Originally created by @psi00 on GitHub (Jul 14, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11418 ### What is the issue? I'm using Msty.app as the UI, which uses Ollama for its backend. Running any model results in a Llama Runner error; `llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed` ### Relevant log output ```shell {"level":50,"time":1752516964568,"pid":10104,"hostname":"Ryzen7","msg":"2025/07/14 19:16:04 routes.go:1232: 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_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:10000 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:3 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:C:\\\\Users\\\\James\\\\AppData\\\\Roaming\\\\Msty\\\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:3 OLLAMA_ORIGINS:[http://localhost http://127.0.0.1 http://0.0.0.0 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://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]\"\n"} {"level":50,"time":1752516964570,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.569+01:00 level=INFO source=images.go:458 msg=\"total blobs: 10\"\n"} {"level":50,"time":1752516964570,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.570+01:00 level=INFO source=images.go:465 msg=\"total unused blobs removed: 0\"\n"} {"level":30,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"[GIN-debug] [WARNING] Creating an Engine instance with the Logger and Recovery middleware already attached.\n\n[GIN-debug] [WARNING] Running in \"debug\" mode. Switch to \"release\" mode in production.\n - using env:\texport GIN_MODE=release\n - using code:\tgin.SetMode(gin.ReleaseMode)\n\n"} {"level":30,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"[GIN-debug] HEAD / --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func1 (5 handlers)\n[GIN-debug] GET / --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func2 (5 handlers)\n[GIN-debug] HEAD /api/version --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func3 (5 handlers)\n[GIN-debug] GET /api/version --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func4 (5 handlers)\n[GIN-debug] POST /api/pull --> github.com/ollama/ollama/server.(*Server).PullHandler-fm (5 handlers)\n[GIN-debug] POST /api/push --> github.com/ollama/ollama/server.(*Server).PushHandler-fm (5 handlers)\n[GIN-debug] HEAD /api/tags --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers)\n[GIN-debug] GET /api/tags --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (5 handlers)\n[GIN-debug] POST /api/show --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (5 handlers)\n[GIN-debug] DELETE /api/delete --> github.com/ollama/ollama/server.(*Server).DeleteHandler-fm (5 handlers)\n[GIN-debug] POST /api/create --> github.com/ollama/ollama/server.(*Server).CreateHandler-fm (5 handlers)\n[GIN-debug] POST /api/blobs/:digest --> github.com/ollama/ollama/server.(*Server).CreateBlobHandler-fm (5 handlers)\n[GIN-debug] HEAD /api/blobs/:digest --> github.com/ollama/ollama/server.(*Server).HeadBlobHandler-fm (5 handlers)\n[GIN-debug] POST /api/copy --> github.com/ollama/ollama/server.(*Server).CopyHandler-fm (5 handlers)\n[GIN-debug] GET /api/ps --> github.com/ollama/ollama/server.(*Server).PsHandler-fm (5 handlers)\n[GIN-debug] POST /api/generate --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (5 handlers)\n[GIN-debug] POST /api/chat --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (5 handlers)\n[GIN-debug] POST /api/embed --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (5 handlers)\n[GIN-debug] POST /api/embeddings --> github.com/ollama/ollama/server.(*Server).EmbeddingsHandler-fm (5 handlers)\n[GIN-debug] POST /v1/chat/completions --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (6 handlers)\n[GIN-debug] POST /v1/completions --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (6 handlers)\n[GIN-debug] POST /v1/embeddings --> github.com/ollama/ollama/server.(*Server).EmbedHandler-fm (6 handlers)\n[GIN-debug] GET /v1/models --> github.com/ollama/ollama/server.(*Server).ListHandler-fm (6 handlers)\n[GIN-debug] GET /v1/models/:model --> github.com/ollama/ollama/server.(*Server).ShowHandler-fm (6 handlers)\n"} {"level":50,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.571+01:00 level=INFO source=routes.go:1299 msg=\"Listening on 127.0.0.1:10000 (version 0.6.6)\"\ntime=2025-07-14T19:16:04.571+01:00 level=INFO source=gpu.go:217 msg=\"looking for compatible GPUs\"\n"} {"level":50,"time":1752516964571,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.571+01:00 level=INFO source=gpu_windows.go:167 msg=packages count=1\ntime=2025-07-14T19:16:04.571+01:00 level=INFO source=gpu_windows.go:214 msg=\"\" package=0 cores=8 efficiency=0 threads=16\n"} {"level":50,"time":1752516964691,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:04.691+01:00 level=INFO source=types.go:130 msg=\"inference compute\" id=GPU-8e0431b3-3727-b8f6-f64e-9303bcf98020 library=cuda variant=v12 compute=8.6 driver=12.8 name=\"NVIDIA GeForce RTX 3060\" total=\"12.0 GiB\" available=\"11.0 GiB\"\n"} {"level":30,"time":1752516964798,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:04 | 200 | 0s | 127.0.0.1 | GET \"/\"\n"} {"level":30,"time":1752516965114,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 | 1.0282ms | 127.0.0.1 | GET \"/api/tags\"\n"} {"level":30,"time":1752516965590,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 | 0s | 127.0.0.1 | GET \"/\"\n"} {"level":30,"time":1752516965707,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 | 0s | 127.0.0.1 | GET \"/\"\n"} {"level":30,"time":1752516965742,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 | 0s | 127.0.0.1 | GET \"/\"\n"} {"level":30,"time":1752516965836,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 | 0s | 127.0.0.1 | GET \"/\"\n"} {"level":30,"time":1752516965839,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:05 | 200 | 1.5081ms | 127.0.0.1 | GET \"/api/tags\"\n"} {"level":50,"time":1752516971235,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.235+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=general.alignment default=32\n"} {"level":50,"time":1752516971265,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.265+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=general.alignment default=32\n"} {"level":50,"time":1752516971277,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.277+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=general.alignment default=32\n"} {"level":50,"time":1752516971278,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.278+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.vision.block_count default=0\n"} {"level":50,"time":1752516971301,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\ntime=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.301+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\n"} {"level":50,"time":1752516971301,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.301+01:00 level=INFO source=sched.go:722 msg=\"new model will fit in available VRAM in single GPU, loading\" model=C:\\Users\\James\\AppData\\Roaming\\Msty\\models\\blobs\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc gpu=GPU-8e0431b3-3727-b8f6-f64e-9303bcf98020 parallel=3 available=11478908928 required=\"1.8 GiB\"\n"} {"level":50,"time":1752516971316,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.315+01:00 level=INFO source=server.go:105 msg=\"system memory\" total=\"15.9 GiB\" free=\"5.0 GiB\" free_swap=\"14.3 GiB\"\ntime=2025-07-14T19:16:11.315+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.vision.block_count default=0\n"} {"level":50,"time":1752516971331,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\n"} {"level":50,"time":1752516971332,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.331+01:00 level=INFO source=server.go:138 msg=offload library=cuda layers.requested=-1 layers.model=29 layers.offload=29 layers.split=\"\" memory.available=\"[10.7 GiB]\" memory.gpu_overhead=\"0 B\" memory.required.full=\"1.8 GiB\" memory.required.partial=\"1.8 GiB\" memory.required.kv=\"168.0 MiB\" memory.required.allocations=\"[1.8 GiB]\" memory.weights.total=\"934.7 MiB\" memory.weights.repeating=\"752.1 MiB\" memory.weights.nonrepeating=\"182.6 MiB\" memory.graph.full=\"299.8 MiB\" memory.graph.partial=\"482.3 MiB\"\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.key_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=WARN source=ggml.go:152 msg=\"key not found\" key=qwen2.attention.value_length default=128\ntime=2025-07-14T19:16:11.331+01:00 level=INFO source=server.go:185 msg=\"enabling flash attention\"\n"} {"level":50,"time":1752516971332,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.331+01:00 level=WARN source=server.go:193 msg=\"kv cache type not supported by model\" type=\"\"\n"} {"level":50,"time":1752516971368,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: loaded meta data with 26 key-value pairs and 339 tensors from C:\\Users\\James\\AppData\\Roaming\\Msty\\models\\blobs\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc (version GGUF V3 (latest))\nllama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.\n"} {"level":50,"time":1752516971368,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 0: general.architecture str = qwen2\nllama_model_loader: - kv 1: general.type str = model\nllama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 1.5B\nllama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen\nllama_model_loader: - kv 4: general.size_label str = 1.5B\nllama_model_loader: - kv 5: qwen2.block_count u32 = 28\nllama_model_loader: - kv 6: qwen2.context_length u32 = 131072\nllama_model_loader: - kv 7: qwen2.embedding_length u32 = 1536\nllama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 8960\nllama_model_loader: - kv 9: qwen2.attention.head_count u32 = 12\nllama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 2\nllama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 10000.000000\nllama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001\nllama_model_loader: - kv 13: general.file_type u32 = 15\nllama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2\nllama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2\n"} {"level":50,"time":1752516971382,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,151936] = [\"!\", \"\\\"\", \"#\", \"$\", \"%\", \"&\", \"'\", ...\n"} {"level":50,"time":1752516971385,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...\n"} {"level":50,"time":1752516971399,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = [\"Ġ Ġ\", \"ĠĠ ĠĠ\", \"i n\", \"Ġ t\",...\nllama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646\nllama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643\nllama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643\nllama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true\nllama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false\nllama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...\nllama_model_loader: - kv 25: general.quantization_version u32 = 2\nllama_model_loader: - type f32: 141 tensors\nllama_model_loader: - type q4_K: 169 tensors\nllama_model_loader: - type q6_K: 29 tensors\n"} {"level":50,"time":1752516971400,"pid":10104,"hostname":"Ryzen7","msg":"print_info: file format = GGUF V3 (latest)\nprint_info: file type = Q4_K - Medium\nprint_info: file size = 1.04 GiB (5.00 BPW) \n"} {"level":50,"time":1752516971496,"pid":10104,"hostname":"Ryzen7","msg":"load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect\n"} {"level":50,"time":1752516971496,"pid":10104,"hostname":"Ryzen7","msg":"load: special tokens cache size = 22\n"} {"level":50,"time":1752516971533,"pid":10104,"hostname":"Ryzen7","msg":"load: token to piece cache size = 0.9310 MB\nprint_info: arch = qwen2\nprint_info: vocab_only = 1\nprint_info: model type = ?B\nprint_info: model params = 1.78 B\nprint_info: general.name = DeepSeek R1 Distill Qwen 1.5B\nprint_info: vocab type = BPE\nprint_info: n_vocab = 151936\nprint_info: n_merges = 151387\nprint_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'\nprint_info: EOS token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOT token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: PAD token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: LF token = 198 'Ċ'\nprint_info: FIM PRE token = 151659 '<|fim_prefix|>'\nprint_info: FIM SUF token = 151661 '<|fim_suffix|>'\n"} {"level":50,"time":1752516971533,"pid":10104,"hostname":"Ryzen7","msg":"print_info: FIM MID token = 151660 '<|fim_middle|>'\nprint_info: FIM PAD token = 151662 '<|fim_pad|>'\nprint_info: FIM REP token = 151663 '<|repo_name|>'\nprint_info: FIM SEP token = 151664 '<|file_sep|>'\nprint_info: EOG token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOG token = 151662 '<|fim_pad|>'\nprint_info: EOG token = 151663 '<|repo_name|>'\nprint_info: EOG token = 151664 '<|file_sep|>'\nprint_info: max token length = 256\nllama_model_load: vocab only - skipping tensors\n"} {"level":50,"time":1752516971543,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.543+01:00 level=INFO source=server.go:405 msg=\"starting llama server\" cmd=\"C:\\\\Users\\\\James\\\\AppData\\\\Roaming\\\\Msty\\\\msty-local.exe runner --model C:\\\\Users\\\\James\\\\AppData\\\\Roaming\\\\Msty\\\\models\\\\blobs\\\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc --ctx-size 6144 --batch-size 512 --n-gpu-layers 29 --threads 8 --flash-attn --no-mmap --parallel 3 --port 59174\"\n"} {"level":50,"time":1752516971548,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.547+01:00 level=INFO source=sched.go:451 msg=\"loaded runners\" count=1\ntime=2025-07-14T19:16:11.547+01:00 level=INFO source=server.go:580 msg=\"waiting for llama runner to start responding\"\n"} {"level":50,"time":1752516971548,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.548+01:00 level=INFO source=server.go:614 msg=\"waiting for server to become available\" status=\"llm server error\"\n"} {"level":50,"time":1752516971582,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.581+01:00 level=INFO source=runner.go:853 msg=\"starting go runner\"\n"} {"level":50,"time":1752516971699,"pid":10104,"hostname":"Ryzen7","msg":"ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no\r\nggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no\r\nggml_cuda_init: found 1 CUDA devices:\r\n Device 0: NVIDIA GeForce RTX 3060, compute capability 8.6, VMM: yes\r\nload_backend: loaded CUDA backend from C:\\Users\\James\\AppData\\Roaming\\Msty\\lib\\ollama\\cuda_v12\\ggml-cuda.dll\n"} {"level":50,"time":1752516971707,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.706+01:00 level=INFO source=ggml.go:109 msg=system CPU.0.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)\n"} {"level":50,"time":1752516971708,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.708+01:00 level=INFO source=runner.go:913 msg=\"Server listening on 127.0.0.1:59174\"\n"} {"level":50,"time":1752516971789,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 3060) - 11242 MiB free\n"} {"level":50,"time":1752516971799,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:11.798+01:00 level=INFO source=server.go:614 msg=\"waiting for server to become available\" status=\"llm server loading model\"\n"} {"level":50,"time":1752516971826,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: loaded meta data with 26 key-value pairs and 339 tensors from C:\\Users\\James\\AppData\\Roaming\\Msty\\models\\blobs\\sha256-aabd4debf0c8f08881923f2c25fc0fdeed24435271c2b3e92c4af36704040dbc (version GGUF V3 (latest))\nllama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.\nllama_model_loader: - kv 0: general.architecture str = qwen2\nllama_model_loader: - kv 1: general.type str = model\nllama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 1.5B\nllama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen\nllama_model_loader: - kv 4: general.size_label str = 1.5B\nllama_model_loader: - kv 5: qwen2.block_count u32 = 28\nllama_model_loader: - kv 6: qwen2.context_length u32 = 131072\nllama_model_loader: - kv 7: qwen2.embedding_length u32 = 1536\nllama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 8960\nllama_model_loader: - kv 9: qwen2.attention.head_count u32 = 12\nllama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 2\nllama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 10000.000000\nllama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001\nllama_model_loader: - kv 13: general.file_type u32 = 15\nllama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2\nllama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2\n"} {"level":50,"time":1752516971840,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,151936] = [\"!\", \"\\\"\", \"#\", \"$\", \"%\", \"&\", \"'\", ...\n"} {"level":50,"time":1752516971844,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...\n"} {"level":50,"time":1752516971859,"pid":10104,"hostname":"Ryzen7","msg":"llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = [\"Ġ Ġ\", \"ĠĠ ĠĠ\", \"i n\", \"Ġ t\",...\nllama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646\nllama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643\nllama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643\nllama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true\nllama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false\nllama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...\nllama_model_loader: - kv 25: general.quantization_version u32 = 2\nllama_model_loader: - type f32: 141 tensors\nllama_model_loader: - type q4_K: 169 tensors\nllama_model_loader: - type q6_K: 29 tensors\nprint_info: file format = GGUF V3 (latest)\nprint_info: file type = Q4_K - Medium\nprint_info: file size = 1.04 GiB (5.00 BPW) \n"} {"level":50,"time":1752516971940,"pid":10104,"hostname":"Ryzen7","msg":"load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect\n"} {"level":50,"time":1752516971940,"pid":10104,"hostname":"Ryzen7","msg":"load: special tokens cache size = 22\n"} {"level":50,"time":1752516971977,"pid":10104,"hostname":"Ryzen7","msg":"load: token to piece cache size = 0.9310 MB\nprint_info: arch = qwen2\nprint_info: vocab_only = 0\nprint_info: n_ctx_train = 131072\nprint_info: n_embd = 1536\nprint_info: n_layer = 28\nprint_info: n_head = 12\nprint_info: n_head_kv = 2\nprint_info: n_rot = 128\nprint_info: n_swa = 0\nprint_info: n_swa_pattern = 1\nprint_info: n_embd_head_k = 128\nprint_info: n_embd_head_v = 128\nprint_info: n_gqa = 6\nprint_info: n_embd_k_gqa = 256\nprint_info: n_embd_v_gqa = 256\nprint_info: f_norm_eps = 0.0e+00\nprint_info: f_norm_rms_eps = 1.0e-06\nprint_info: f_clamp_kqv = 0.0e+00\nprint_info: f_max_alibi_bias = 0.0e+00\nprint_info: f_logit_scale = 0.0e+00\nprint_info: f_attn_scale = 0.0e+00\nprint_info: n_ff = 8960\nprint_info: n_expert = 0\nprint_info: n_expert_used = 0\nprint_info: causal attn = 1\nprint_info: pooling type = 0\nprint_info: rope type = 2\nprint_info: rope scaling = linear\nprint_info: freq_base_train = 10000.0\nprint_info: freq_scale_train = 1\nprint_info: n_ctx_orig_yarn = 131072\nprint_info: rope_finetuned = unknown\nprint_info: ssm_d_conv = 0\nprint_info: ssm_d_inner = 0\nprint_info: ssm_d_state = 0\nprint_info: ssm_dt_rank = 0\nprint_info: ssm_dt_b_c_rms = 0\nprint_info: model type = 1.5B\nprint_info: model params = 1.78 B\nprint_info: general.name = DeepSeek R1 Distill Qwen 1.5B\nprint_info: vocab type = BPE\nprint_info: n_vocab = 151936\nprint_info: n_merges = 151387\nprint_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'\nprint_info: EOS token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOT token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: PAD token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: LF token = 198 'Ċ'\nprint_info: FIM PRE token = 151659 '<|fim_prefix|>'\nprint_info: FIM SUF token = 151661 '<|fim_suffix|>'\nprint_info: FIM MID token = 151660 '<|fim_middle|>'\nprint_info: FIM PAD token = 151662 '<|fim_pad|>'\nprint_info: FIM REP token = 151663 '<|repo_name|>'\nprint_info: FIM SEP token = 151664 '<|file_sep|>'\nprint_info: EOG token = 151643 '<｜end▁of▁sentence｜>'\nprint_info: EOG token = 151662 '<|fim_pad|>'\nprint_info: EOG token = 151663 '<|repo_name|>'\nprint_info: EOG token = 151664 '<|file_sep|>'\nprint_info: max token length = 256\nload_tensors: loading model tensors, this can take a while... (mmap = false)\n"} {"level":50,"time":1752516971982,"pid":10104,"hostname":"Ryzen7","msg":"load_tensors: offloading 28 repeating layers to GPU\nload_tensors: offloading output layer to GPU\nload_tensors: offloaded 29/29 layers to GPU\nload_tensors: CPU model buffer size = 125.19 MiB\nload_tensors: CUDA0 model buffer size = 934.70 MiB\n"} {"level":50,"time":1752516972202,"pid":10104,"hostname":"Ryzen7","msg":"llama_context: constructing llama_context\nllama_context: n_seq_max = 3\nllama_context: n_ctx = 6144\nllama_context: n_ctx_per_seq = 2048\nllama_context: n_batch = 1536\nllama_context: n_ubatch = 512\nllama_context: causal_attn = 1\nllama_context: flash_attn = 1\nllama_context: freq_base = 10000.0\nllama_context: freq_scale = 1\nllama_context: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized\n"} {"level":50,"time":1752516972203,"pid":10104,"hostname":"Ryzen7","msg":"llama_context: CUDA_Host output buffer size = 1.76 MiB\ninit: kv_size = 6144, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1\n"} {"level":50,"time":1752516972209,"pid":10104,"hostname":"Ryzen7","msg":"init: CUDA0 KV buffer size = 168.00 MiB\nllama_context: KV self size = 168.00 MiB, K (f16): 84.00 MiB, V (f16): 84.00 MiB\n"} {"level":50,"time":1752516972217,"pid":10104,"hostname":"Ryzen7","msg":"D:/a/llama.cpp/llama.cpp/ggml/src/ggml.c:1777: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed\r\n"} {"level":50,"time":1752516972501,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:12.500+01:00 level=INFO source=server.go:614 msg=\"waiting for server to become available\" status=\"llm server not responding\"\n"} {"level":50,"time":1752516972565,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:12.565+01:00 level=ERROR source=server.go:449 msg=\"llama runner terminated\" error=\"exit status 0xc0000409\"\n"} {"level":50,"time":1752516972751,"pid":10104,"hostname":"Ryzen7","msg":"time=2025-07-14T19:16:12.750+01:00 level=ERROR source=sched.go:457 msg=\"error loading llama server\" error=\"llama runner process has terminated: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed\"\n"} {"level":30,"time":1752516972752,"pid":10104,"hostname":"Ryzen7","msg":"[GIN] 2025/07/14 - 19:16:12 | 500 | 1.5308332s | 127.0.0.1 | POST \"/api/chat\"\n"} ``` ### OS Windows ### GPU AMD ### CPU AMD ### Ollama version 0.6.6
GiteaMirror added the bug label 2026-05-04 18:36:05 -05:00
GiteaMirror commented 2026-05-04 18:36:07 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 14, 2025):

{"level":50,"time":1752516972217,"pid":10104,"hostname":"Ryzen7","msg":"D:/a/llama.cpp/llama.cpp/ggml/src/ggml.c:1777: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed\r\n"}

#9509

Upgrade ollama.

<!-- gh-comment-id:3070551068 --> @rick-github commented on GitHub (Jul 14, 2025): ``` {"level":50,"time":1752516972217,"pid":10104,"hostname":"Ryzen7","msg":"D:/a/llama.cpp/llama.cpp/ggml/src/ggml.c:1777: GGML_ASSERT(tensor->op == GGML_OP_UNARY) failed\r\n"} ``` #9509 [Upgrade](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-can-i-upgrade-ollama) ollama.
GiteaMirror commented 2026-05-04 18:36:08 -05:00
Author
Owner
Copy Link

@psi00 commented on GitHub (Jul 14, 2025):

Thanks. I'll have to let the developers of Msty know since there's no way to update Ollama within their UI

<!-- gh-comment-id:3070586523 --> @psi00 commented on GitHub (Jul 14, 2025): Thanks. I'll have to let the developers of Msty know since there's no way to update Ollama within their UI
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#69596
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?