github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11168] Ollama is not using available resources CPU + GPU #7366

New Issue
Closed
opened 2026-04-12 19:25:39 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 19:25:39 -05:00
Owner
Copy Link

Originally created by @faci2000 on GitHub (Jun 23, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11168

What is the issue?

I'm trying to run ollama on AWS EC2 g4dn.12xlarge instance, but it is not reaching the full potential.

Using the deepseek-r1:32b model, it is reaching usage of GPUs just around 30%:
Image

Average performance looks like here:
Image

I've noticed that during the response

docker-inspect.txt

ollama is fully using only one core of the CPU:
Image

Inside docker container:
Image

Outside:
Image

Nevertheless, the docker container is not restriced in any way, and have possibility to use whole CPU.
Image

I've tried playing with parameters like: num_thread or tried setting the env variables regarding multicore usage, but without success.

logs.log
lscpu.txt
nvidia-smi.txt

Relevant log output

time=2025-06-18T13:06:18.358Z level=INFO source=server.go:630 msg="llama runner started in 6.78 seconds"
[GIN] 2025/06/18 - 13:07:41 | 200 |         1m35s |      172.20.0.2 | POST     "/api/chat"
[GIN] 2025/06/18 - 13:08:03 | 200 |        14m21s |      172.20.0.2 | POST     "/api/chat"
[GIN] 2025/06/18 - 13:08:06 | 200 | 24.671084422s |      172.20.0.2 | POST     "/api/chat"
[GIN] 2025/06/18 - 13:08:08 | 200 |      35.566µs |      172.20.0.2 | GET      "/api/version"
[GIN] 2025/06/18 - 13:08:13 | 200 |      35.206µs |      172.20.0.2 | GET      "/api/version"
[GIN] 2025/06/18 - 13:08:15 | 200 |      33.833µs |      172.20.0.2 | GET      "/api/version"
[GIN] 2025/06/18 - 13:08:41 | 200 |      35.448µs |      172.20.0.2 | GET      "/api/version"
[GIN] 2025/06/18 - 13:08:55 | 200 |     877.804µs |      172.20.0.2 | GET      "/api/tags"
[GIN] 2025/06/18 - 13:08:55 | 200 |      29.454µs |      172.20.0.2 | GET      "/api/ps"
[GIN] 2025/06/18 - 13:08:59 | 500 | 52.956108807s |      172.20.0.2 | POST     "/api/chat"
time=2025-06-18T13:09:00.836Z level=INFO source=routes.go:1206 msg="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:4096 OLLAMA_DEBUG:INFO 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:[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: http_proxy: https_proxy: no_proxy:]"
time=2025-06-18T13:09:00.839Z level=INFO source=images.go:463 msg="total blobs: 34"
time=2025-06-18T13:09:00.839Z level=INFO source=images.go:470 msg="total unused blobs removed: 0"
time=2025-06-18T13:09:00.840Z level=INFO source=routes.go:1259 msg="Listening on [::]:11434 (version 0.8.0)"
time=2025-06-18T13:09:00.840Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-76077695-d38e-6877-07c4-37549f5b7862 library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB"
time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-7fa2e651-fc04-8ece-2a8e-65348cc2a35b library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB"
time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-cd9b52f1-7931-d509-4ad7-fd1f543682b1 library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB"
time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-9321216a-55b1-30c4-eb0b-c8bad3a0b84c library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB"
[GIN] 2025/06/18 - 13:09:40 | 200 |     1.21188ms |      172.20.0.2 | GET      "/api/tags"
[GIN] 2025/06/18 - 13:09:40 | 200 |     103.587µs |      172.20.0.2 | GET      "/api/ps"
[GIN] 2025/06/18 - 13:09:41 | 200 |        65.1µs |      172.20.0.2 | GET      "/api/version"
time=2025-06-18T13:10:44.654Z level=INFO source=sched.go:804 msg="new model will fit in available VRAM, loading" model=/root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 library=cuda parallel=2 required="26.7 GiB"
time=2025-06-18T13:10:45.709Z level=INFO source=server.go:135 msg="system memory" total="186.7 GiB" free="183.0 GiB" free_swap="0 B"
time=2025-06-18T13:10:45.709Z level=INFO source=server.go:168 msg=offload library=cuda layers.requested=-1 layers.model=65 layers.offload=65 layers.split=17,16,16,16 memory.available="[14.5 GiB 14.5 GiB 14.5 GiB 14.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="26.7 GiB" memory.required.partial="26.7 GiB" memory.required.kv="2.0 GiB" memory.required.allocations="[7.1 GiB 6.5 GiB 6.5 GiB 6.5 GiB]" memory.weights.total="18.1 GiB" memory.weights.repeating="17.5 GiB" memory.weights.nonrepeating="609.1 MiB" memory.graph.full="916.1 MiB" memory.graph.partial="916.1 MiB"
llama_model_loader: loaded meta data with 26 key-value pairs and 771 tensors from /root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = qwen2
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 32B
llama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv   4:                         general.size_label str              = 32B
llama_model_loader: - kv   5:                          qwen2.block_count u32              = 64
llama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072
llama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 5120
llama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 27648
llama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 40
llama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  13:                          general.file_type u32              = 15
llama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = deepseek-r1-qwen
llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  321 tensors
llama_model_loader: - type q4_K:  385 tensors
llama_model_loader: - type q6_K:   65 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 18.48 GiB (4.85 BPW) 
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch             = qwen2
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 32.76 B
print_info: general.name     = DeepSeek R1 Distill Qwen 32B
print_info: vocab type       = BPE
print_info: n_vocab          = 152064
print_info: n_merges         = 151387
print_info: BOS token        = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token        = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token        = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token        = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2025-06-18T13:10:45.987Z level=INFO source=server.go:431 msg="starting llama server" cmd="/usr/bin/ollama runner --model /root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 --ctx-size 8192 --batch-size 512 --n-gpu-layers 65 --threads 24 --parallel 2 --tensor-split 17,16,16,16 --port 41815"
time=2025-06-18T13:10:45.987Z level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-06-18T13:10:45.987Z level=INFO source=server.go:591 msg="waiting for llama runner to start responding"
time=2025-06-18T13:10:45.988Z level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server not responding"
time=2025-06-18T13:10:46.001Z level=INFO source=runner.go:815 msg="starting go runner"
load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-skylakex.so
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 4 CUDA devices:
  Device 0: Tesla T4, compute capability 7.5, VMM: yes
  Device 1: Tesla T4, compute capability 7.5, VMM: yes
  Device 2: Tesla T4, compute capability 7.5, VMM: yes
  Device 3: Tesla T4, compute capability 7.5, VMM: yes
load_backend: loaded CUDA backend from /usr/lib/ollama/cuda_v12/libggml-cuda.so
time=2025-06-18T13:10:46.665Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.LLAMAFILE=1 CPU.1.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 CUDA.1.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.1.USE_GRAPHS=1 CUDA.1.PEER_MAX_BATCH_SIZE=128 CUDA.2.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.2.USE_GRAPHS=1 CUDA.2.PEER_MAX_BATCH_SIZE=128 CUDA.3.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.3.USE_GRAPHS=1 CUDA.3.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)
time=2025-06-18T13:10:46.665Z level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:41815"
time=2025-06-18T13:10:46.740Z level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server loading model"
llama_model_load_from_file_impl: using device CUDA0 (Tesla T4) - 14825 MiB free
llama_model_load_from_file_impl: using device CUDA1 (Tesla T4) - 14825 MiB free
llama_model_load_from_file_impl: using device CUDA2 (Tesla T4) - 14825 MiB free
llama_model_load_from_file_impl: using device CUDA3 (Tesla T4) - 14825 MiB free
llama_model_loader: loaded meta data with 26 key-value pairs and 771 tensors from /root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = qwen2
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 32B
llama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv   4:                         general.size_label str              = 32B
llama_model_loader: - kv   5:                          qwen2.block_count u32              = 64
llama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072
llama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 5120
llama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 27648
llama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 40
llama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  13:                          general.file_type u32              = 15
llama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = deepseek-r1-qwen
llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  321 tensors
llama_model_loader: - type q4_K:  385 tensors
llama_model_loader: - type q6_K:   65 tensors

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.8.0

Originally created by @faci2000 on GitHub (Jun 23, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11168 ### What is the issue? I'm trying to run ollama on AWS EC2 g4dn.12xlarge instance, but it is not reaching the full potential. Using the deepseek-r1:32b model, it is reaching usage of GPUs just around 30%: ![Image](https://github.com/user-attachments/assets/1ad113cf-f31e-4290-88a6-254beb37fee1) Average performance looks like here: ![Image](https://github.com/user-attachments/assets/1b72eca8-255d-496a-99b1-8ce0ff8b3382) I've noticed that during the response [docker-inspect.txt](https://github.com/user-attachments/files/20864617/docker-inspect.txt) ollama is fully using only one core of the CPU: ![Image](https://github.com/user-attachments/assets/9a7bbba2-4e68-4afd-b492-94a8a607adab) Inside docker container: ![Image](https://github.com/user-attachments/assets/e7617e1b-acf8-4884-9a63-65832ee3adcf) Outside: ![Image](https://github.com/user-attachments/assets/1777f1d9-a6ac-44fa-b050-59673d4561f5) Nevertheless, the docker container is not restriced in any way, and have possibility to use whole CPU. ![Image](https://github.com/user-attachments/assets/abe4ffd8-e8a2-4473-aa64-f419d422b4af) I've tried playing with parameters like: num_thread or tried setting the env variables regarding multicore usage, but without success. [logs.log](https://github.com/user-attachments/files/20864618/logs.log) [lscpu.txt](https://github.com/user-attachments/files/20864619/lscpu.txt) [nvidia-smi.txt](https://github.com/user-attachments/files/20864616/nvidia-smi.txt) ### Relevant log output ```shell time=2025-06-18T13:06:18.358Z level=INFO source=server.go:630 msg="llama runner started in 6.78 seconds" [GIN] 2025/06/18 - 13:07:41 | 200 | 1m35s | 172.20.0.2 | POST "/api/chat" [GIN] 2025/06/18 - 13:08:03 | 200 | 14m21s | 172.20.0.2 | POST "/api/chat" [GIN] 2025/06/18 - 13:08:06 | 200 | 24.671084422s | 172.20.0.2 | POST "/api/chat" [GIN] 2025/06/18 - 13:08:08 | 200 | 35.566µs | 172.20.0.2 | GET "/api/version" [GIN] 2025/06/18 - 13:08:13 | 200 | 35.206µs | 172.20.0.2 | GET "/api/version" [GIN] 2025/06/18 - 13:08:15 | 200 | 33.833µs | 172.20.0.2 | GET "/api/version" [GIN] 2025/06/18 - 13:08:41 | 200 | 35.448µs | 172.20.0.2 | GET "/api/version" [GIN] 2025/06/18 - 13:08:55 | 200 | 877.804µs | 172.20.0.2 | GET "/api/tags" [GIN] 2025/06/18 - 13:08:55 | 200 | 29.454µs | 172.20.0.2 | GET "/api/ps" [GIN] 2025/06/18 - 13:08:59 | 500 | 52.956108807s | 172.20.0.2 | POST "/api/chat" time=2025-06-18T13:09:00.836Z level=INFO source=routes.go:1206 msg="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:4096 OLLAMA_DEBUG:INFO 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:[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: http_proxy: https_proxy: no_proxy:]" time=2025-06-18T13:09:00.839Z level=INFO source=images.go:463 msg="total blobs: 34" time=2025-06-18T13:09:00.839Z level=INFO source=images.go:470 msg="total unused blobs removed: 0" time=2025-06-18T13:09:00.840Z level=INFO source=routes.go:1259 msg="Listening on [::]:11434 (version 0.8.0)" time=2025-06-18T13:09:00.840Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-76077695-d38e-6877-07c4-37549f5b7862 library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB" time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-7fa2e651-fc04-8ece-2a8e-65348cc2a35b library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB" time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-cd9b52f1-7931-d509-4ad7-fd1f543682b1 library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB" time=2025-06-18T13:09:01.833Z level=INFO source=types.go:130 msg="inference compute" id=GPU-9321216a-55b1-30c4-eb0b-c8bad3a0b84c library=cuda variant=v12 compute=7.5 driver=12.2 name="Tesla T4" total="14.6 GiB" available="14.5 GiB" [GIN] 2025/06/18 - 13:09:40 | 200 | 1.21188ms | 172.20.0.2 | GET "/api/tags" [GIN] 2025/06/18 - 13:09:40 | 200 | 103.587µs | 172.20.0.2 | GET "/api/ps" [GIN] 2025/06/18 - 13:09:41 | 200 | 65.1µs | 172.20.0.2 | GET "/api/version" time=2025-06-18T13:10:44.654Z level=INFO source=sched.go:804 msg="new model will fit in available VRAM, loading" model=/root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 library=cuda parallel=2 required="26.7 GiB" time=2025-06-18T13:10:45.709Z level=INFO source=server.go:135 msg="system memory" total="186.7 GiB" free="183.0 GiB" free_swap="0 B" time=2025-06-18T13:10:45.709Z level=INFO source=server.go:168 msg=offload library=cuda layers.requested=-1 layers.model=65 layers.offload=65 layers.split=17,16,16,16 memory.available="[14.5 GiB 14.5 GiB 14.5 GiB 14.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="26.7 GiB" memory.required.partial="26.7 GiB" memory.required.kv="2.0 GiB" memory.required.allocations="[7.1 GiB 6.5 GiB 6.5 GiB 6.5 GiB]" memory.weights.total="18.1 GiB" memory.weights.repeating="17.5 GiB" memory.weights.nonrepeating="609.1 MiB" memory.graph.full="916.1 MiB" memory.graph.partial="916.1 MiB" llama_model_loader: loaded meta data with 26 key-value pairs and 771 tensors from /root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 32B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 32B llama_model_loader: - kv 5: qwen2.block_count u32 = 64 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 27648 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = deepseek-r1-qwen llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 321 tensors llama_model_loader: - type q4_K: 385 tensors llama_model_loader: - type q6_K: 65 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 18.48 GiB (4.85 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 32.76 B print_info: general.name = DeepSeek R1 Distill Qwen 32B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2025-06-18T13:10:45.987Z level=INFO source=server.go:431 msg="starting llama server" cmd="/usr/bin/ollama runner --model /root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 --ctx-size 8192 --batch-size 512 --n-gpu-layers 65 --threads 24 --parallel 2 --tensor-split 17,16,16,16 --port 41815" time=2025-06-18T13:10:45.987Z level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-06-18T13:10:45.987Z level=INFO source=server.go:591 msg="waiting for llama runner to start responding" time=2025-06-18T13:10:45.988Z level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server not responding" time=2025-06-18T13:10:46.001Z level=INFO source=runner.go:815 msg="starting go runner" load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-skylakex.so ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 4 CUDA devices: Device 0: Tesla T4, compute capability 7.5, VMM: yes Device 1: Tesla T4, compute capability 7.5, VMM: yes Device 2: Tesla T4, compute capability 7.5, VMM: yes Device 3: Tesla T4, compute capability 7.5, VMM: yes load_backend: loaded CUDA backend from /usr/lib/ollama/cuda_v12/libggml-cuda.so time=2025-06-18T13:10:46.665Z level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.LLAMAFILE=1 CPU.1.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 CUDA.1.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.1.USE_GRAPHS=1 CUDA.1.PEER_MAX_BATCH_SIZE=128 CUDA.2.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.2.USE_GRAPHS=1 CUDA.2.PEER_MAX_BATCH_SIZE=128 CUDA.3.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.3.USE_GRAPHS=1 CUDA.3.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc) time=2025-06-18T13:10:46.665Z level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:41815" time=2025-06-18T13:10:46.740Z level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server loading model" llama_model_load_from_file_impl: using device CUDA0 (Tesla T4) - 14825 MiB free llama_model_load_from_file_impl: using device CUDA1 (Tesla T4) - 14825 MiB free llama_model_load_from_file_impl: using device CUDA2 (Tesla T4) - 14825 MiB free llama_model_load_from_file_impl: using device CUDA3 (Tesla T4) - 14825 MiB free llama_model_loader: loaded meta data with 26 key-value pairs and 771 tensors from /root/.ollama/models/blobs/sha256-6150cb382311b69f09cc0f9a1b69fc029cbd742b66bb8ec531aa5ecf5c613e93 (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 32B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 32B llama_model_loader: - kv 5: qwen2.block_count u32 = 64 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 27648 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = deepseek-r1-qwen llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 321 tensors llama_model_loader: - type q4_K: 385 tensors llama_model_loader: - type q6_K: 65 tensors ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.8.0
GiteaMirror added the bug label 2026-04-12 19:25:39 -05:00
GiteaMirror commented 2026-04-12 19:25:41 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 23, 2025):

https://github.com/ollama/ollama/issues/7648#issuecomment-2473561990

<!-- gh-comment-id:2996197410 --> @rick-github commented on GitHub (Jun 23, 2025): https://github.com/ollama/ollama/issues/7648#issuecomment-2473561990
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#7366
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?