github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-23 06:01:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #12722] 100% GPU Model Runs 100% CPU #8442

New Issue
Closed
opened 2026-04-12 21:07:14 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-04-12 21:07:14 -05:00
Owner
Copy Link

Originally created by @Panican-Whyasker on GitHub (Oct 21, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12722

What is the issue?

Running a 3.2-GB model with Ollama set to the lowest 4k context in order to fit in 6 GB of VRAM available.

ollama ps assures that model runs 100% GPU; however, neither VRAM nor GPU is used at all.

Image

Image

Image

server.log shows that the nVidia GPU was visible to Ollama, and 5.8 GB of VRAM was available (strangely, the server.log is still being written, long after I have stopped the model):

Relevant log output

time=2025-10-21T13:15:09.345+02:00 level=INFO source=routes.go:1511 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_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://127.0.0.1:11434 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:C:\\Users\\gyordano\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 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_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-10-21T13:15:09.476+02:00 level=INFO source=images.go:522 msg="total blobs: 123"
time=2025-10-21T13:15:09.483+02:00 level=INFO source=images.go:529 msg="total unused blobs removed: 0"
time=2025-10-21T13:15:09.487+02:00 level=INFO source=routes.go:1564 msg="Listening on 127.0.0.1:11434 (version 0.12.6)"
time=2025-10-21T13:15:09.488+02:00 level=INFO source=runner.go:80 msg="discovering available GPUs..."
time=2025-10-21T13:15:12.992+02:00 level=INFO source=types.go:112 msg="inference compute" id=GPU-6da04c1e-b5be-4fd5-6109-a2d7c38483ca library=CUDA compute=7.5 name=CUDA0 description="Quadro RTX 3000" libdirs=ollama,cuda_v12 driver=12.8 pci_id=01:00.0 type=discrete total="6.0 GiB" available="5.8 GiB"
time=2025-10-21T13:15:12.992+02:00 level=INFO source=routes.go:1605 msg="entering low vram mode" "total vram"="6.0 GiB" threshold="20.0 GiB"
[GIN] 2025/10/21 - 13:15:18 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/10/21 - 13:15:18 | 200 |     63.5234ms |       127.0.0.1 | POST     "/api/show"
llama_model_loader: loaded meta data with 36 key-value pairs and 196 tensors from C:\Users\gyordano\.ollama\models\blobs\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 (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              = phi3
llama_model_loader: - kv   1:              phi3.rope.scaling.attn_factor f32              = 1.190238
llama_model_loader: - kv   2:                               general.type str              = model
llama_model_loader: - kv   3:                               general.name str              = Phi 4 Mini Reasoning
llama_model_loader: - kv   4:                           general.finetune str              = reasoning
llama_model_loader: - kv   5:                           general.basename str              = Phi-4
llama_model_loader: - kv   6:                         general.size_label str              = mini
llama_model_loader: - kv   7:                            general.license str              = mit
llama_model_loader: - kv   8:                       general.license.link str              = https://huggingface.co/microsoft/Phi-...
llama_model_loader: - kv   9:                               general.tags arr[str,4]       = ["nlp", "math", "code", "text-generat...
llama_model_loader: - kv  10:                          general.languages arr[str,1]       = ["en"]
llama_model_loader: - kv  11:                        phi3.context_length u32              = 131072
llama_model_loader: - kv  12:  phi3.rope.scaling.original_context_length u32              = 4096
llama_model_loader: - kv  13:                      phi3.embedding_length u32              = 3072
llama_model_loader: - kv  14:                   phi3.feed_forward_length u32              = 8192
llama_model_loader: - kv  15:                           phi3.block_count u32              = 32
llama_model_loader: - kv  16:                  phi3.attention.head_count u32              = 24
llama_model_loader: - kv  17:               phi3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  18:      phi3.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  19:                  phi3.rope.dimension_count u32              = 96
llama_model_loader: - kv  20:                        phi3.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  21:              phi3.attention.sliding_window u32              = 262144
llama_model_loader: - kv  22:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  23:                         tokenizer.ggml.pre str              = gpt-4o
llama_model_loader: - kv  24:                      tokenizer.ggml.tokens arr[str,200064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  25:                  tokenizer.ggml.token_type arr[i32,200064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  26:                      tokenizer.ggml.merges arr[str,199742]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "e r", ...
llama_model_loader: - kv  27:                tokenizer.ggml.bos_token_id u32              = 199999
llama_model_loader: - kv  28:                tokenizer.ggml.eos_token_id u32              = 199999
llama_model_loader: - kv  29:            tokenizer.ggml.unknown_token_id u32              = 199999
llama_model_loader: - kv  30:            tokenizer.ggml.padding_token_id u32              = 199999
llama_model_loader: - kv  31:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  32:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  33:                    tokenizer.chat_template str              = {{ '<|system|>Your name is Phi, an AI...
llama_model_loader: - kv  34:               general.quantization_version u32              = 2
llama_model_loader: - kv  35:                          general.file_type u32              = 15
llama_model_loader: - type  f32:   67 tensors
llama_model_loader: - type  f16:   32 tensors
llama_model_loader: - type q4_K:   80 tensors
llama_model_loader: - type q6_K:   17 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 2.93 GiB (6.56 BPW) 
load: printing all EOG tokens:
load:   - 199999 ('<|endoftext|>')
load:   - 200020 ('<|end|>')
load: special tokens cache size = 12
load: token to piece cache size = 1.3333 MB
print_info: arch             = phi3
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 3.84 B
print_info: general.name     = Phi 4 Mini Reasoning
print_info: vocab type       = BPE
print_info: n_vocab          = 200064
print_info: n_merges         = 199742
print_info: BOS token        = 199999 '<|endoftext|>'
print_info: EOS token        = 199999 '<|endoftext|>'
print_info: EOT token        = 200020 '<|end|>'
print_info: UNK token        = 199999 '<|endoftext|>'
print_info: PAD token        = 199999 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: EOG token        = 199999 '<|endoftext|>'
print_info: EOG token        = 200020 '<|end|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2025-10-21T13:15:19.230+02:00 level=INFO source=cpu_windows.go:139 msg=packages count=1
time=2025-10-21T13:15:19.230+02:00 level=INFO source=cpu_windows.go:186 msg="" package=0 cores=8 efficiency=0 threads=16
time=2025-10-21T13:15:19.231+02:00 level=INFO source=server.go:400 msg="starting runner" cmd="C:\\Users\\gyordano\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\gyordano\\.ollama\\models\\blobs\\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 --port 59242"
time=2025-10-21T13:15:19.537+02:00 level=INFO source=cpu_windows.go:139 msg=packages count=1
time=2025-10-21T13:15:19.537+02:00 level=INFO source=cpu_windows.go:186 msg="" package=0 cores=8 efficiency=0 threads=16
time=2025-10-21T13:15:19.537+02:00 level=INFO source=server.go:505 msg="system memory" total="127.7 GiB" free="105.8 GiB" free_swap="120.8 GiB"
time=2025-10-21T13:15:19.537+02:00 level=INFO source=memory.go:36 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=C:\Users\gyordano\.ollama\models\blobs\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 library=CUDA parallel=1 required="4.2 GiB" gpus=1
time=2025-10-21T13:15:19.537+02:00 level=INFO source=server.go:545 msg=offload library=CUDA layers.requested=-1 layers.model=33 layers.offload=33 layers.split=[33] memory.available="[5.8 GiB]" memory.gpu_overhead="0 B" memory.required.full="4.2 GiB" memory.required.partial="4.2 GiB" memory.required.kv="512.0 MiB" memory.required.allocations="[4.2 GiB]" memory.weights.total="2.9 GiB" memory.weights.repeating="2.5 GiB" memory.weights.nonrepeating="480.8 MiB" memory.graph.full="256.0 MiB" memory.graph.partial="256.0 MiB"
time=2025-10-21T13:15:19.667+02:00 level=INFO source=runner.go:893 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\gyordano\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
  Device 0: Quadro RTX 3000, compute capability 7.5, VMM: yes, ID: GPU-6da04c1e-b5be-4fd5-6109-a2d7c38483ca
load_backend: loaded CUDA backend from C:\Users\gyordano\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
time=2025-10-21T13:15:19.804+02:00 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.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-10-21T13:15:19.806+02:00 level=INFO source=runner.go:929 msg="Server listening on 127.0.0.1:59242"
time=2025-10-21T13:15:19.814+02:00 level=INFO source=runner.go:828 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:4096 KvCacheType: NumThreads:8 GPULayers:33[ID:GPU-6da04c1e-b5be-4fd5-6109-a2d7c38483ca Layers:33(0..32)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-21T13:15:19.814+02:00 level=INFO source=server.go:1272 msg="waiting for llama runner to start responding"
time=2025-10-21T13:15:19.814+02:00 level=INFO source=server.go:1306 msg="waiting for server to become available" status="llm server loading model"
ggml_backend_cuda_device_get_memory utilizing NVML memory reporting free: 6244270080 total: 6442450944
llama_model_load_from_file_impl: using device CUDA0 (Quadro RTX 3000) (0000:01:00.0) - 5955 MiB free
llama_model_loader: loaded meta data with 36 key-value pairs and 196 tensors from C:\Users\gyordano\.ollama\models\blobs\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 (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              = phi3
llama_model_loader: - kv   1:              phi3.rope.scaling.attn_factor f32              = 1.190238
llama_model_loader: - kv   2:                               general.type str              = model
llama_model_loader: - kv   3:                               general.name str              = Phi 4 Mini Reasoning
llama_model_loader: - kv   4:                           general.finetune str              = reasoning
llama_model_loader: - kv   5:                           general.basename str              = Phi-4
llama_model_loader: - kv   6:                         general.size_label str              = mini
llama_model_loader: - kv   7:                            general.license str              = mit
llama_model_loader: - kv   8:                       general.license.link str              = https://huggingface.co/microsoft/Phi-...
llama_model_loader: - kv   9:                               general.tags arr[str,4]       = ["nlp", "math", "code", "text-generat...
llama_model_loader: - kv  10:                          general.languages arr[str,1]       = ["en"]
llama_model_loader: - kv  11:                        phi3.context_length u32              = 131072
llama_model_loader: - kv  12:  phi3.rope.scaling.original_context_length u32              = 4096
llama_model_loader: - kv  13:                      phi3.embedding_length u32              = 3072
llama_model_loader: - kv  14:                   phi3.feed_forward_length u32              = 8192
llama_model_loader: - kv  15:                           phi3.block_count u32              = 32
llama_model_loader: - kv  16:                  phi3.attention.head_count u32              = 24
llama_model_loader: - kv  17:               phi3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  18:      phi3.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  19:                  phi3.rope.dimension_count u32              = 96
llama_model_loader: - kv  20:                        phi3.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  21:              phi3.attention.sliding_window u32              = 262144
llama_model_loader: - kv  22:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  23:                         tokenizer.ggml.pre str              = gpt-4o
llama_model_loader: - kv  24:                      tokenizer.ggml.tokens arr[str,200064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  25:                  tokenizer.ggml.token_type arr[i32,200064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  26:                      tokenizer.ggml.merges arr[str,199742]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "e r", ...
llama_model_loader: - kv  27:                tokenizer.ggml.bos_token_id u32              = 199999
llama_model_loader: - kv  28:                tokenizer.ggml.eos_token_id u32              = 199999
llama_model_loader: - kv  29:            tokenizer.ggml.unknown_token_id u32              = 199999
llama_model_loader: - kv  30:            tokenizer.ggml.padding_token_id u32              = 199999
llama_model_loader: - kv  31:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  32:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  33:                    tokenizer.chat_template str              = {{ '<|system|>Your name is Phi, an AI...
llama_model_loader: - kv  34:               general.quantization_version u32              = 2
llama_model_loader: - kv  35:                          general.file_type u32              = 15
llama_model_loader: - type  f32:   67 tensors
llama_model_loader: - type  f16:   32 tensors
llama_model_loader: - type q4_K:   80 tensors
llama_model_loader: - type q6_K:   17 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 2.93 GiB (6.56 BPW) 
load_hparams: Phi SWA is currently disabled - results might be suboptimal for some models (see https://github.com/ggml-org/llama.cpp/pull/13676)
load: printing all EOG tokens:
load:   - 199999 ('<|endoftext|>')
load:   - 200020 ('<|end|>')
load: special tokens cache size = 12
load: token to piece cache size = 1.3333 MB
print_info: arch             = phi3
print_info: vocab_only       = 0
print_info: n_ctx_train      = 131072
print_info: n_embd           = 3072
print_info: n_layer          = 32
print_info: n_head           = 24
print_info: n_head_kv        = 8
print_info: n_rot            = 96
print_info: n_swa            = 0
print_info: is_swa_any       = 0
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 3
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-05
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 8192
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 10000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 4096
print_info: rope_finetuned   = unknown
print_info: model type       = 3B
print_info: model params     = 3.84 B
print_info: general.name     = Phi 4 Mini Reasoning
print_info: vocab type       = BPE
print_info: n_vocab          = 200064
print_info: n_merges         = 199742
print_info: BOS token        = 199999 '<|endoftext|>'
print_info: EOS token        = 199999 '<|endoftext|>'
print_info: EOT token        = 200020 '<|end|>'
print_info: UNK token        = 199999 '<|endoftext|>'
print_info: PAD token        = 199999 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: EOG token        = 199999 '<|endoftext|>'
print_info: EOG token        = 200020 '<|end|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 32 repeating layers to GPU
load_tensors: offloading output layer to GPU
load_tensors: offloaded 33/33 layers to GPU
load_tensors:        CUDA0 model buffer size =  2998.57 MiB
load_tensors:          CPU model buffer size =   480.81 MiB
llama_init_from_model: model default pooling_type is [0], but [-1] was specified
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 4096
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = disabled
llama_context: kv_unified    = false
llama_context: freq_base     = 10000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_per_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context:  CUDA_Host  output buffer size =     0.77 MiB
llama_kv_cache:      CUDA0 KV buffer size =   512.00 MiB
llama_kv_cache: size =  512.00 MiB (  4096 cells,  32 layers,  1/1 seqs), K (f16):  256.00 MiB, V (f16):  256.00 MiB
llama_context:      CUDA0 compute buffer size =   402.75 MiB
llama_context:  CUDA_Host compute buffer size =    18.01 MiB
llama_context: graph nodes  = 1094
llama_context: graph splits = 2
time=2025-10-21T13:15:21.570+02:00 level=INFO source=server.go:1310 msg="llama runner started in 2.34 seconds"
time=2025-10-21T13:15:21.570+02:00 level=INFO source=sched.go:482 msg="loaded runners" count=1
time=2025-10-21T13:15:21.570+02:00 level=INFO source=server.go:1272 msg="waiting for llama runner to start responding"
time=2025-10-21T13:15:21.570+02:00 level=INFO source=server.go:1310 msg="llama runner started in 2.34 seconds"
[GIN] 2025/10/21 - 13:15:21 | 200 |    3.4289956s |       127.0.0.1 | POST     "/api/generate"
[GIN] 2025/10/21 - 13:15:59 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/10/21 - 13:15:59 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2025/10/21 - 13:16:37 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/10/21 - 13:16:37 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2025/10/21 - 13:16:39 | 200 |   10.7630465s |       127.0.0.1 | POST     "/api/chat"
[GIN] 2025/10/21 - 13:18:52 | 200 |    12.346345s |       127.0.0.1 | POST     "/api/chat"
[GIN] 2025/10/21 - 13:19:06 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/10/21 - 13:19:06 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2025/10/21 - 13:19:22 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/10/21 - 13:19:22 | 200 |      5.8093ms |       127.0.0.1 | POST     "/api/generate"
[GIN] 2025/10/21 - 13:29:50 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/10/21 - 13:34:42 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/10/21 - 13:34:42 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2025/10/21 - 13:35:22 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/10/21 - 13:35:22 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/10/21 - 13:35:23 | 200 |     11.3883ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/10/21 - 13:35:23 | 200 |     26.4085ms |       127.0.0.1 | POST     "/api/show"

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.12.6

Originally created by @Panican-Whyasker on GitHub (Oct 21, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12722 ### What is the issue? Running a 3.2-GB model with Ollama set to the lowest 4k context in order to fit in 6 GB of VRAM available. ollama ps assures that model runs 100% GPU; however, neither VRAM nor GPU is used at all. ![Image](https://github.com/user-attachments/assets/eadf3747-5f1d-47f4-b550-54451c9d98d4) ![Image](https://github.com/user-attachments/assets/3e0fe215-0b0d-4878-9c5a-33724acce2ab) ![Image](https://github.com/user-attachments/assets/10140d74-6ae0-4821-af77-b3fb635ede0f) server.log shows that the nVidia GPU was visible to Ollama, and 5.8 GB of VRAM was available (strangely, the server.log is still being written, long after I have stopped the model): ### Relevant log output ```shell time=2025-10-21T13:15:09.345+02:00 level=INFO source=routes.go:1511 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_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://127.0.0.1:11434 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:C:\\Users\\gyordano\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 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_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-10-21T13:15:09.476+02:00 level=INFO source=images.go:522 msg="total blobs: 123" time=2025-10-21T13:15:09.483+02:00 level=INFO source=images.go:529 msg="total unused blobs removed: 0" time=2025-10-21T13:15:09.487+02:00 level=INFO source=routes.go:1564 msg="Listening on 127.0.0.1:11434 (version 0.12.6)" time=2025-10-21T13:15:09.488+02:00 level=INFO source=runner.go:80 msg="discovering available GPUs..." time=2025-10-21T13:15:12.992+02:00 level=INFO source=types.go:112 msg="inference compute" id=GPU-6da04c1e-b5be-4fd5-6109-a2d7c38483ca library=CUDA compute=7.5 name=CUDA0 description="Quadro RTX 3000" libdirs=ollama,cuda_v12 driver=12.8 pci_id=01:00.0 type=discrete total="6.0 GiB" available="5.8 GiB" time=2025-10-21T13:15:12.992+02:00 level=INFO source=routes.go:1605 msg="entering low vram mode" "total vram"="6.0 GiB" threshold="20.0 GiB" [GIN] 2025/10/21 - 13:15:18 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/10/21 - 13:15:18 | 200 | 63.5234ms | 127.0.0.1 | POST "/api/show" llama_model_loader: loaded meta data with 36 key-value pairs and 196 tensors from C:\Users\gyordano\.ollama\models\blobs\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 (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 = phi3 llama_model_loader: - kv 1: phi3.rope.scaling.attn_factor f32 = 1.190238 llama_model_loader: - kv 2: general.type str = model llama_model_loader: - kv 3: general.name str = Phi 4 Mini Reasoning llama_model_loader: - kv 4: general.finetune str = reasoning llama_model_loader: - kv 5: general.basename str = Phi-4 llama_model_loader: - kv 6: general.size_label str = mini llama_model_loader: - kv 7: general.license str = mit llama_model_loader: - kv 8: general.license.link str = https://huggingface.co/microsoft/Phi-... llama_model_loader: - kv 9: general.tags arr[str,4] = ["nlp", "math", "code", "text-generat... llama_model_loader: - kv 10: general.languages arr[str,1] = ["en"] llama_model_loader: - kv 11: phi3.context_length u32 = 131072 llama_model_loader: - kv 12: phi3.rope.scaling.original_context_length u32 = 4096 llama_model_loader: - kv 13: phi3.embedding_length u32 = 3072 llama_model_loader: - kv 14: phi3.feed_forward_length u32 = 8192 llama_model_loader: - kv 15: phi3.block_count u32 = 32 llama_model_loader: - kv 16: phi3.attention.head_count u32 = 24 llama_model_loader: - kv 17: phi3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 18: phi3.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 19: phi3.rope.dimension_count u32 = 96 llama_model_loader: - kv 20: phi3.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 21: phi3.attention.sliding_window u32 = 262144 llama_model_loader: - kv 22: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 23: tokenizer.ggml.pre str = gpt-4o llama_model_loader: - kv 24: tokenizer.ggml.tokens arr[str,200064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 25: tokenizer.ggml.token_type arr[i32,200064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 26: tokenizer.ggml.merges arr[str,199742] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "e r", ... llama_model_loader: - kv 27: tokenizer.ggml.bos_token_id u32 = 199999 llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 199999 llama_model_loader: - kv 29: tokenizer.ggml.unknown_token_id u32 = 199999 llama_model_loader: - kv 30: tokenizer.ggml.padding_token_id u32 = 199999 llama_model_loader: - kv 31: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 32: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 33: tokenizer.chat_template str = {{ '<|system|>Your name is Phi, an AI... llama_model_loader: - kv 34: general.quantization_version u32 = 2 llama_model_loader: - kv 35: general.file_type u32 = 15 llama_model_loader: - type f32: 67 tensors llama_model_loader: - type f16: 32 tensors llama_model_loader: - type q4_K: 80 tensors llama_model_loader: - type q6_K: 17 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 2.93 GiB (6.56 BPW) load: printing all EOG tokens: load: - 199999 ('<|endoftext|>') load: - 200020 ('<|end|>') load: special tokens cache size = 12 load: token to piece cache size = 1.3333 MB print_info: arch = phi3 print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 3.84 B print_info: general.name = Phi 4 Mini Reasoning print_info: vocab type = BPE print_info: n_vocab = 200064 print_info: n_merges = 199742 print_info: BOS token = 199999 '<|endoftext|>' print_info: EOS token = 199999 '<|endoftext|>' print_info: EOT token = 200020 '<|end|>' print_info: UNK token = 199999 '<|endoftext|>' print_info: PAD token = 199999 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: EOG token = 199999 '<|endoftext|>' print_info: EOG token = 200020 '<|end|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2025-10-21T13:15:19.230+02:00 level=INFO source=cpu_windows.go:139 msg=packages count=1 time=2025-10-21T13:15:19.230+02:00 level=INFO source=cpu_windows.go:186 msg="" package=0 cores=8 efficiency=0 threads=16 time=2025-10-21T13:15:19.231+02:00 level=INFO source=server.go:400 msg="starting runner" cmd="C:\\Users\\gyordano\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\gyordano\\.ollama\\models\\blobs\\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 --port 59242" time=2025-10-21T13:15:19.537+02:00 level=INFO source=cpu_windows.go:139 msg=packages count=1 time=2025-10-21T13:15:19.537+02:00 level=INFO source=cpu_windows.go:186 msg="" package=0 cores=8 efficiency=0 threads=16 time=2025-10-21T13:15:19.537+02:00 level=INFO source=server.go:505 msg="system memory" total="127.7 GiB" free="105.8 GiB" free_swap="120.8 GiB" time=2025-10-21T13:15:19.537+02:00 level=INFO source=memory.go:36 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=C:\Users\gyordano\.ollama\models\blobs\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 library=CUDA parallel=1 required="4.2 GiB" gpus=1 time=2025-10-21T13:15:19.537+02:00 level=INFO source=server.go:545 msg=offload library=CUDA layers.requested=-1 layers.model=33 layers.offload=33 layers.split=[33] memory.available="[5.8 GiB]" memory.gpu_overhead="0 B" memory.required.full="4.2 GiB" memory.required.partial="4.2 GiB" memory.required.kv="512.0 MiB" memory.required.allocations="[4.2 GiB]" memory.weights.total="2.9 GiB" memory.weights.repeating="2.5 GiB" memory.weights.nonrepeating="480.8 MiB" memory.graph.full="256.0 MiB" memory.graph.partial="256.0 MiB" time=2025-10-21T13:15:19.667+02:00 level=INFO source=runner.go:893 msg="starting go runner" load_backend: loaded CPU backend from C:\Users\gyordano\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: Quadro RTX 3000, compute capability 7.5, VMM: yes, ID: GPU-6da04c1e-b5be-4fd5-6109-a2d7c38483ca load_backend: loaded CUDA backend from C:\Users\gyordano\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll time=2025-10-21T13:15:19.804+02:00 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.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-10-21T13:15:19.806+02:00 level=INFO source=runner.go:929 msg="Server listening on 127.0.0.1:59242" time=2025-10-21T13:15:19.814+02:00 level=INFO source=runner.go:828 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:4096 KvCacheType: NumThreads:8 GPULayers:33[ID:GPU-6da04c1e-b5be-4fd5-6109-a2d7c38483ca Layers:33(0..32)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-21T13:15:19.814+02:00 level=INFO source=server.go:1272 msg="waiting for llama runner to start responding" time=2025-10-21T13:15:19.814+02:00 level=INFO source=server.go:1306 msg="waiting for server to become available" status="llm server loading model" ggml_backend_cuda_device_get_memory utilizing NVML memory reporting free: 6244270080 total: 6442450944 llama_model_load_from_file_impl: using device CUDA0 (Quadro RTX 3000) (0000:01:00.0) - 5955 MiB free llama_model_loader: loaded meta data with 36 key-value pairs and 196 tensors from C:\Users\gyordano\.ollama\models\blobs\sha256-f4dd2368e6c32725dc1c5c5548ae9ee2724d6a79052952eb50b65e26288022c4 (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 = phi3 llama_model_loader: - kv 1: phi3.rope.scaling.attn_factor f32 = 1.190238 llama_model_loader: - kv 2: general.type str = model llama_model_loader: - kv 3: general.name str = Phi 4 Mini Reasoning llama_model_loader: - kv 4: general.finetune str = reasoning llama_model_loader: - kv 5: general.basename str = Phi-4 llama_model_loader: - kv 6: general.size_label str = mini llama_model_loader: - kv 7: general.license str = mit llama_model_loader: - kv 8: general.license.link str = https://huggingface.co/microsoft/Phi-... llama_model_loader: - kv 9: general.tags arr[str,4] = ["nlp", "math", "code", "text-generat... llama_model_loader: - kv 10: general.languages arr[str,1] = ["en"] llama_model_loader: - kv 11: phi3.context_length u32 = 131072 llama_model_loader: - kv 12: phi3.rope.scaling.original_context_length u32 = 4096 llama_model_loader: - kv 13: phi3.embedding_length u32 = 3072 llama_model_loader: - kv 14: phi3.feed_forward_length u32 = 8192 llama_model_loader: - kv 15: phi3.block_count u32 = 32 llama_model_loader: - kv 16: phi3.attention.head_count u32 = 24 llama_model_loader: - kv 17: phi3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 18: phi3.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 19: phi3.rope.dimension_count u32 = 96 llama_model_loader: - kv 20: phi3.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 21: phi3.attention.sliding_window u32 = 262144 llama_model_loader: - kv 22: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 23: tokenizer.ggml.pre str = gpt-4o llama_model_loader: - kv 24: tokenizer.ggml.tokens arr[str,200064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 25: tokenizer.ggml.token_type arr[i32,200064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 26: tokenizer.ggml.merges arr[str,199742] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "e r", ... llama_model_loader: - kv 27: tokenizer.ggml.bos_token_id u32 = 199999 llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 199999 llama_model_loader: - kv 29: tokenizer.ggml.unknown_token_id u32 = 199999 llama_model_loader: - kv 30: tokenizer.ggml.padding_token_id u32 = 199999 llama_model_loader: - kv 31: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 32: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 33: tokenizer.chat_template str = {{ '<|system|>Your name is Phi, an AI... llama_model_loader: - kv 34: general.quantization_version u32 = 2 llama_model_loader: - kv 35: general.file_type u32 = 15 llama_model_loader: - type f32: 67 tensors llama_model_loader: - type f16: 32 tensors llama_model_loader: - type q4_K: 80 tensors llama_model_loader: - type q6_K: 17 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 2.93 GiB (6.56 BPW) load_hparams: Phi SWA is currently disabled - results might be suboptimal for some models (see https://github.com/ggml-org/llama.cpp/pull/13676) load: printing all EOG tokens: load: - 199999 ('<|endoftext|>') load: - 200020 ('<|end|>') load: special tokens cache size = 12 load: token to piece cache size = 1.3333 MB print_info: arch = phi3 print_info: vocab_only = 0 print_info: n_ctx_train = 131072 print_info: n_embd = 3072 print_info: n_layer = 32 print_info: n_head = 24 print_info: n_head_kv = 8 print_info: n_rot = 96 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 3 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 8192 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 1 print_info: pooling type = 0 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 10000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 4096 print_info: rope_finetuned = unknown print_info: model type = 3B print_info: model params = 3.84 B print_info: general.name = Phi 4 Mini Reasoning print_info: vocab type = BPE print_info: n_vocab = 200064 print_info: n_merges = 199742 print_info: BOS token = 199999 '<|endoftext|>' print_info: EOS token = 199999 '<|endoftext|>' print_info: EOT token = 200020 '<|end|>' print_info: UNK token = 199999 '<|endoftext|>' print_info: PAD token = 199999 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: EOG token = 199999 '<|endoftext|>' print_info: EOG token = 200020 '<|end|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 32 repeating layers to GPU load_tensors: offloading output layer to GPU load_tensors: offloaded 33/33 layers to GPU load_tensors: CUDA0 model buffer size = 2998.57 MiB load_tensors: CPU model buffer size = 480.81 MiB llama_init_from_model: model default pooling_type is [0], but [-1] was specified llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = disabled llama_context: kv_unified = false llama_context: freq_base = 10000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_context: CUDA_Host output buffer size = 0.77 MiB llama_kv_cache: CUDA0 KV buffer size = 512.00 MiB llama_kv_cache: size = 512.00 MiB ( 4096 cells, 32 layers, 1/1 seqs), K (f16): 256.00 MiB, V (f16): 256.00 MiB llama_context: CUDA0 compute buffer size = 402.75 MiB llama_context: CUDA_Host compute buffer size = 18.01 MiB llama_context: graph nodes = 1094 llama_context: graph splits = 2 time=2025-10-21T13:15:21.570+02:00 level=INFO source=server.go:1310 msg="llama runner started in 2.34 seconds" time=2025-10-21T13:15:21.570+02:00 level=INFO source=sched.go:482 msg="loaded runners" count=1 time=2025-10-21T13:15:21.570+02:00 level=INFO source=server.go:1272 msg="waiting for llama runner to start responding" time=2025-10-21T13:15:21.570+02:00 level=INFO source=server.go:1310 msg="llama runner started in 2.34 seconds" [GIN] 2025/10/21 - 13:15:21 | 200 | 3.4289956s | 127.0.0.1 | POST "/api/generate" [GIN] 2025/10/21 - 13:15:59 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/10/21 - 13:15:59 | 200 | 0s | 127.0.0.1 | GET "/api/ps" [GIN] 2025/10/21 - 13:16:37 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/10/21 - 13:16:37 | 200 | 0s | 127.0.0.1 | GET "/api/ps" [GIN] 2025/10/21 - 13:16:39 | 200 | 10.7630465s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/10/21 - 13:18:52 | 200 | 12.346345s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/10/21 - 13:19:06 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/10/21 - 13:19:06 | 200 | 0s | 127.0.0.1 | GET "/api/ps" [GIN] 2025/10/21 - 13:19:22 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/10/21 - 13:19:22 | 200 | 5.8093ms | 127.0.0.1 | POST "/api/generate" [GIN] 2025/10/21 - 13:29:50 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/10/21 - 13:34:42 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/10/21 - 13:34:42 | 200 | 0s | 127.0.0.1 | GET "/api/ps" [GIN] 2025/10/21 - 13:35:22 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/10/21 - 13:35:22 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/10/21 - 13:35:23 | 200 | 11.3883ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/10/21 - 13:35:23 | 200 | 26.4085ms | 127.0.0.1 | POST "/api/show" ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.12.6
GiteaMirror added the bug label 2026-04-12 21:07:14 -05:00
GiteaMirror commented 2026-04-12 21:07:17 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 21, 2025):

What does nvidia-smi show?

<!-- gh-comment-id:3426251696 --> @rick-github commented on GitHub (Oct 21, 2025): What does `nvidia-smi` show?
GiteaMirror commented 2026-04-12 21:07:19 -05:00
Author
Owner
Copy Link

@Panican-Whyasker commented on GitHub (Oct 21, 2025):

@rick-github here it goes:

PS D:\JORO\Docs\Sci\A.I> nvidia-smi
Tue Oct 21 14:09:15 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 573.44 Driver Version: 573.44 CUDA Version: 12.8 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+
| 0 Quadro RTX 3000 WDDM | 00000000:01:00.0 Off | N/A |
| N/A 55C P8 5W / 65W | 0MiB / 6144MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|====================================================================
+-----------------------------------------------------------------------------------------+

<!-- gh-comment-id:3426282276 --> @Panican-Whyasker commented on GitHub (Oct 21, 2025): @rick-github here it goes: PS D:\JORO\Docs\Sci\A.I> nvidia-smi Tue Oct 21 14:09:15 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 573.44 Driver Version: 573.44 CUDA Version: 12.8 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+ | 0 Quadro RTX 3000 WDDM | 00000000:01:00.0 Off | N/A | | N/A 55C P8 5W / 65W | 0MiB / 6144MiB | 0% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |==================================================================== +-----------------------------------------------------------------------------------------+
GiteaMirror commented 2026-04-12 21:07:21 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 21, 2025):

What does ollama ps show?

<!-- gh-comment-id:3426285901 --> @rick-github commented on GitHub (Oct 21, 2025): What does `ollama ps` show?
GiteaMirror commented 2026-04-12 21:07:23 -05:00
Author
Owner
Copy Link

@Panican-Whyasker commented on GitHub (Oct 21, 2025):

PS C:\Windows\System32\WindowsPowerShell\v1.0> ollama ps
NAME ID SIZE PROCESSOR CONTEXT UNTIL
PS C:\Windows\System32\WindowsPowerShell\v1.0>

<!-- gh-comment-id:3426290124 --> @Panican-Whyasker commented on GitHub (Oct 21, 2025): PS C:\Windows\System32\WindowsPowerShell\v1.0> ollama ps NAME ID SIZE PROCESSOR CONTEXT UNTIL PS C:\Windows\System32\WindowsPowerShell\v1.0>
GiteaMirror commented 2026-04-12 21:07:24 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 21, 2025):

What does the following show:

ollama run phi4-mini-reasoning ""
ollama ps
nvidia-smi
<!-- gh-comment-id:3426297154 --> @rick-github commented on GitHub (Oct 21, 2025): What does the following show: ``` ollama run phi4-mini-reasoning "" ollama ps nvidia-smi ```
GiteaMirror commented 2026-04-12 21:07:25 -05:00
Author
Owner
Copy Link

@Panican-Whyasker commented on GitHub (Oct 21, 2025):

Image

Image

<!-- gh-comment-id:3426308187 --> @Panican-Whyasker commented on GitHub (Oct 21, 2025): ![Image](https://github.com/user-attachments/assets/ec77a342-a430-456c-9650-84d2e65f202b) ![Image](https://github.com/user-attachments/assets/703105d8-4470-4b69-ad41-74b94044555e)
GiteaMirror commented 2026-04-12 21:07:26 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 21, 2025):

The GPU is being used. Enable CUDA monitoring as described here.

<!-- gh-comment-id:3426315587 --> @rick-github commented on GitHub (Oct 21, 2025): The GPU is being used. Enable CUDA monitoring as described [here](https://michaelceber.medium.com/gpu-monitoring-on-windows-10-for-machine-learning-cuda-41088de86d65).
GiteaMirror commented 2026-04-12 21:07:26 -05:00
Author
Owner
Copy Link

@Panican-Whyasker commented on GitHub (Oct 21, 2025):

Okay, it seems that the bug is not in Ollama but in Windows Task Manager. After re-starting it, it now shows the nVidia GPU VRAM and engines use correctly:

Image

<!-- gh-comment-id:3426336387 --> @Panican-Whyasker commented on GitHub (Oct 21, 2025): Okay, it seems that the bug is not in Ollama but in Windows Task Manager. After re-starting it, it now shows the nVidia GPU VRAM and engines use correctly: ![Image](https://github.com/user-attachments/assets/3f34ea1c-c489-47e5-b51a-af44840e9cb8)
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#8442
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?