github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-15 12:48:05 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #10978] Model running in CPU, even though GPU is recognized #7235

New Issue
Closed
opened 2026-04-12 19:15:39 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 19:15:39 -05:00
Owner
Copy Link

Originally created by @CastilloDel on GitHub (Jun 5, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10978

What is the issue?

I expect the models to run in GPU. I have tried with two rather small models qwen2.5-coder and tinydolphin, which should fit in the memory of my GPU, but for some reason they use my CPU. If I run nvidia-smi I can see that my gpu is completely idle. I also checked I get the same speed when just using CUDA_VISIBLE_DEVICES="" (then I can see in the logs that ollama doesn't detect the GPU).

Relevant log output

time=2025-06-05T11:50:27.957+02:00 level=INFO source=routes.go:1234 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PR
time=2025-06-05T11:50:27.958+02:00 level=INFO source=images.go:479 msg="total blobs: 11"
time=2025-06-05T11:50:27.958+02:00 level=INFO source=images.go:486 msg="total unused blobs removed: 0"
time=2025-06-05T11:50:27.958+02:00 level=INFO source=routes.go:1287 msg="Listening on 127.0.0.1:11434 (version 0.9.0)"
time=2025-06-05T11:50:27.958+02:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-06-05T11:50:28.232+02:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-6b310bc7-c8da-0676-7802-6c2ce1453721 library=cuda variant=v12 compute=8.6 driver=12.8 name="NVIDI
[GIN] 2025/06/05 - 11:50:37 | 200 |       54.25µs |       127.0.0.1 | HEAD     "/"
[GIN] 2025/06/05 - 11:50:37 | 200 |   10.115158ms |       127.0.0.1 | POST     "/api/show"
time=2025-06-05T11:50:37.359+02:00 level=INFO source=sched.go:788 msg="new model will fit in available VRAM in single GPU, loading" model=/home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a
time=2025-06-05T11:50:37.517+02:00 level=INFO source=server.go:135 msg="system memory" total="31.0 GiB" free="24.3 GiB" free_swap="32.0 GiB"
time=2025-06-05T11:50:37.517+02:00 level=INFO source=server.go:168 msg=offload library=cuda layers.requested=-1 layers.model=23 layers.offload=23 layers.split="" memory.available="[7.5 GiB]" mem
llama_model_loader: loaded meta data with 21 key-value pairs and 201 tensors from /home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a65557d1daddaa16e26a1dd9b66dc6a52ae94260a3f0078348 (versi
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = llama
llama_model_loader: - kv   1:                               general.name str              = LLaMA v2
llama_model_loader: - kv   2:                       llama.context_length u32              = 4096
llama_model_loader: - kv   3:                     llama.embedding_length u32              = 2048
llama_model_loader: - kv   4:                          llama.block_count u32              = 22
llama_model_loader: - kv   5:                  llama.feed_forward_length u32              = 5632
llama_model_loader: - kv   6:                 llama.rope.dimension_count u32              = 64
llama_model_loader: - kv   7:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv   8:              llama.attention.head_count_kv u32              = 4
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  10:                       llama.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  11:                          general.file_type u32              = 2
llama_model_loader: - kv  12:                       tokenizer.ggml.model str              = llama
llama_model_loader: - kv  13:                      tokenizer.ggml.tokens arr[str,32002]   = ["<unk>", "<s>", "</s>", "<0x00>", "<...
llama_model_loader: - kv  14:                      tokenizer.ggml.scores arr[f32,32002]   = [0.000000, 0.000000, 0.000000, 0.0000...
llama_model_loader: - kv  15:                  tokenizer.ggml.token_type arr[i32,32002]   = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ...
llama_model_loader: - kv  16:                tokenizer.ggml.bos_token_id u32              = 1
llama_model_loader: - kv  17:                tokenizer.ggml.eos_token_id u32              = 32000
llama_model_loader: - kv  18:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  19:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  20:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   45 tensors
llama_model_loader: - type q4_0:  155 tensors
llama_model_loader: - type q6_K:    1 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_0
print_info: file size   = 606.53 MiB (4.63 BPW)
load: control-looking token:  32000 '<|im_end|>' was not control-type; this is probably a bug in the model. its type will be overridden
load: special tokens cache size = 5
load: token to piece cache size = 0.1684 MB
print_info: arch             = llama
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 1.10 B
print_info: general.name     = LLaMA v2
print_info: vocab type       = SPM
print_info: n_vocab          = 32002
print_info: n_merges         = 0
print_info: BOS token        = 1 '<s>'
print_info: EOS token        = 32000 '<|im_end|>'
print_info: EOT token        = 32000 '<|im_end|>'
print_info: UNK token        = 0 '<unk>'
print_info: LF token         = 13 '<0x0A>'
print_info: EOG token        = 32000 '<|im_end|>'
print_info: max token length = 48
llama_model_load: vocab only - skipping tensors
time=2025-06-05T11:50:37.539+02:00 level=INFO source=server.go:431 msg="starting llama server" cmd="/usr/bin/ollama runner --model /home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a65557d1
time=2025-06-05T11:50:37.539+02:00 level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-06-05T11:50:37.539+02:00 level=INFO source=server.go:591 msg="waiting for llama runner to start responding"
time=2025-06-05T11:50:37.540+02:00 level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server not responding"
time=2025-06-05T11:50:37.550+02:00 level=INFO source=runner.go:815 msg="starting go runner"
load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-alderlake.so
time=2025-06-05T11:50:37.553+02:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMA
time=2025-06-05T11:50:37.553+02:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:44639"
llama_model_loader: loaded meta data with 21 key-value pairs and 201 tensors from /home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a65557d1daddaa16e26a1dd9b66dc6a52ae94260a3f0078348 (versi
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = llama
llama_model_loader: - kv   1:                               general.name str              = LLaMA v2
llama_model_loader: - kv   2:                       llama.context_length u32              = 4096
llama_model_loader: - kv   3:                     llama.embedding_length u32              = 2048
llama_model_loader: - kv   4:                          llama.block_count u32              = 22
llama_model_loader: - kv   5:                  llama.feed_forward_length u32              = 5632
llama_model_loader: - kv   6:                 llama.rope.dimension_count u32              = 64
llama_model_loader: - kv   7:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv   8:              llama.attention.head_count_kv u32              = 4
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  10:                       llama.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  11:                          general.file_type u32              = 2
llama_model_loader: - kv  12:                       tokenizer.ggml.model str              = llama
llama_model_loader: - kv  13:                      tokenizer.ggml.tokens arr[str,32002]   = ["<unk>", "<s>", "</s>", "<0x00>", "<...
llama_model_loader: - kv  14:                      tokenizer.ggml.scores arr[f32,32002]   = [0.000000, 0.000000, 0.000000, 0.0000...
llama_model_loader: - kv  15:                  tokenizer.ggml.token_type arr[i32,32002]   = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ...
llama_model_loader: - kv  16:                tokenizer.ggml.bos_token_id u32              = 1
llama_model_loader: - kv  17:                tokenizer.ggml.eos_token_id u32              = 32000
llama_model_loader: - kv  18:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  19:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  20:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   45 tensors
llama_model_loader: - type q4_0:  155 tensors
llama_model_loader: - type q6_K:    1 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_0
print_info: file size   = 606.53 MiB (4.63 BPW)
load: control-looking token:  32000 '<|im_end|>' was not control-type; this is probably a bug in the model. its type will be overridden
load: special tokens cache size = 5
load: token to piece cache size = 0.1684 MB
print_info: arch             = llama
print_info: vocab_only       = 0
print_info: n_ctx_train      = 4096
print_info: n_embd           = 2048
print_info: n_layer          = 22
print_info: n_head           = 32
print_info: n_head_kv        = 4
print_info: n_rot            = 64
print_info: n_swa            = 0
print_info: n_swa_pattern    = 1
print_info: n_embd_head_k    = 64
print_info: n_embd_head_v    = 64
print_info: n_gqa            = 8
print_info: n_embd_k_gqa     = 256
print_info: n_embd_v_gqa     = 256
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             = 5632
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        = 0
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: ssm_d_conv       = 0
print_info: ssm_d_inner      = 0
print_info: ssm_d_state      = 0
print_info: ssm_dt_rank      = 0
print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 1B
print_info: model params     = 1.10 B
print_info: general.name     = LLaMA v2
print_info: vocab type       = SPM
print_info: n_vocab          = 32002
print_info: n_merges         = 0
print_info: BOS token        = 1 '<s>'
print_info: EOS token        = 32000 '<|im_end|>'
print_info: EOT token        = 32000 '<|im_end|>'
print_info: UNK token        = 0 '<unk>'
print_info: LF token         = 13 '<0x0A>'
print_info: EOG token        = 32000 '<|im_end|>'
print_info: max token length = 48
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors:   CPU_Mapped model buffer size =   606.53 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 2
llama_context: n_ctx         = 8192
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch       = 1024
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = 0
llama_context: freq_base     = 10000.0
llama_context: freq_scale    = 1
llama_context:        CPU  output buffer size =     0.26 MiB
llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 22, can_shift = 1, padding = 32
llama_kv_cache_unified:        CPU KV buffer size =   176.00 MiB
llama_kv_cache_unified: KV self size  =  176.00 MiB, K (f16):   88.00 MiB, V (f16):   88.00 MiB
llama_context:        CPU compute buffer size =   544.01 MiB
llama_context: graph nodes  = 754
llama_context: graph splits = 1
time=2025-06-05T11:50:37.791+02:00 level=INFO source=server.go:630 msg="llama runner started in 0.25 seconds"
[GIN] 2025/06/05 - 11:50:37 | 200 |  614.135872ms |       127.0.0.1 | POST     "/api/generate"
[GIN] 2025/06/05 - 11:50:53 | 200 |  8.678716289s |       127.0.0.1 | POST     "/api/chat"
[GIN] 2025/06/05 - 11:51:16 | 200 | 15.028160535s |       127.0.0.1 | POST     "/api/chat"

[GIN] 2025/06/05 - 11:53:54 | 200 |      73.113µs |       127.0.0.1 | GET      "/api/version"

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @CastilloDel on GitHub (Jun 5, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10978 ### What is the issue? I expect the models to run in GPU. I have tried with two rather small models `qwen2.5-coder` and `tinydolphin`, which should fit in the memory of my GPU, but for some reason they use my CPU. If I run `nvidia-smi` I can see that my gpu is completely idle. I also checked I get the same speed when just using CUDA_VISIBLE_DEVICES="" (then I can see in the logs that ollama doesn't detect the GPU). ### Relevant log output ```shell time=2025-06-05T11:50:27.957+02:00 level=INFO source=routes.go:1234 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PR time=2025-06-05T11:50:27.958+02:00 level=INFO source=images.go:479 msg="total blobs: 11" time=2025-06-05T11:50:27.958+02:00 level=INFO source=images.go:486 msg="total unused blobs removed: 0" time=2025-06-05T11:50:27.958+02:00 level=INFO source=routes.go:1287 msg="Listening on 127.0.0.1:11434 (version 0.9.0)" time=2025-06-05T11:50:27.958+02:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-06-05T11:50:28.232+02:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-6b310bc7-c8da-0676-7802-6c2ce1453721 library=cuda variant=v12 compute=8.6 driver=12.8 name="NVIDI [GIN] 2025/06/05 - 11:50:37 | 200 | 54.25µs | 127.0.0.1 | HEAD "/" [GIN] 2025/06/05 - 11:50:37 | 200 | 10.115158ms | 127.0.0.1 | POST "/api/show" time=2025-06-05T11:50:37.359+02:00 level=INFO source=sched.go:788 msg="new model will fit in available VRAM in single GPU, loading" model=/home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a time=2025-06-05T11:50:37.517+02:00 level=INFO source=server.go:135 msg="system memory" total="31.0 GiB" free="24.3 GiB" free_swap="32.0 GiB" time=2025-06-05T11:50:37.517+02:00 level=INFO source=server.go:168 msg=offload library=cuda layers.requested=-1 layers.model=23 layers.offload=23 layers.split="" memory.available="[7.5 GiB]" mem llama_model_loader: loaded meta data with 21 key-value pairs and 201 tensors from /home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a65557d1daddaa16e26a1dd9b66dc6a52ae94260a3f0078348 (versi llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = llama llama_model_loader: - kv 1: general.name str = LLaMA v2 llama_model_loader: - kv 2: llama.context_length u32 = 4096 llama_model_loader: - kv 3: llama.embedding_length u32 = 2048 llama_model_loader: - kv 4: llama.block_count u32 = 22 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 5632 llama_model_loader: - kv 6: llama.rope.dimension_count u32 = 64 llama_model_loader: - kv 7: llama.attention.head_count u32 = 32 llama_model_loader: - kv 8: llama.attention.head_count_kv u32 = 4 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: llama.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 11: general.file_type u32 = 2 llama_model_loader: - kv 12: tokenizer.ggml.model str = llama llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,32002] = ["<unk>", "<s>", "</s>", "<0x00>", "<... llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,32002] = [0.000000, 0.000000, 0.000000, 0.0000... llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,32002] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... llama_model_loader: - kv 16: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 17: tokenizer.ggml.eos_token_id u32 = 32000 llama_model_loader: - kv 18: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 19: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 20: general.quantization_version u32 = 2 llama_model_loader: - type f32: 45 tensors llama_model_loader: - type q4_0: 155 tensors llama_model_loader: - type q6_K: 1 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_0 print_info: file size = 606.53 MiB (4.63 BPW) load: control-looking token: 32000 '<|im_end|>' was not control-type; this is probably a bug in the model. its type will be overridden load: special tokens cache size = 5 load: token to piece cache size = 0.1684 MB print_info: arch = llama print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 1.10 B print_info: general.name = LLaMA v2 print_info: vocab type = SPM print_info: n_vocab = 32002 print_info: n_merges = 0 print_info: BOS token = 1 '<s>' print_info: EOS token = 32000 '<|im_end|>' print_info: EOT token = 32000 '<|im_end|>' print_info: UNK token = 0 '<unk>' print_info: LF token = 13 '<0x0A>' print_info: EOG token = 32000 '<|im_end|>' print_info: max token length = 48 llama_model_load: vocab only - skipping tensors time=2025-06-05T11:50:37.539+02:00 level=INFO source=server.go:431 msg="starting llama server" cmd="/usr/bin/ollama runner --model /home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a65557d1 time=2025-06-05T11:50:37.539+02:00 level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-06-05T11:50:37.539+02:00 level=INFO source=server.go:591 msg="waiting for llama runner to start responding" time=2025-06-05T11:50:37.540+02:00 level=INFO source=server.go:625 msg="waiting for server to become available" status="llm server not responding" time=2025-06-05T11:50:37.550+02:00 level=INFO source=runner.go:815 msg="starting go runner" load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-alderlake.so time=2025-06-05T11:50:37.553+02:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMA time=2025-06-05T11:50:37.553+02:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:44639" llama_model_loader: loaded meta data with 21 key-value pairs and 201 tensors from /home/daniel/.ollama/models/blobs/sha256-5996bfb2c06d79a65557d1daddaa16e26a1dd9b66dc6a52ae94260a3f0078348 (versi llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = llama llama_model_loader: - kv 1: general.name str = LLaMA v2 llama_model_loader: - kv 2: llama.context_length u32 = 4096 llama_model_loader: - kv 3: llama.embedding_length u32 = 2048 llama_model_loader: - kv 4: llama.block_count u32 = 22 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 5632 llama_model_loader: - kv 6: llama.rope.dimension_count u32 = 64 llama_model_loader: - kv 7: llama.attention.head_count u32 = 32 llama_model_loader: - kv 8: llama.attention.head_count_kv u32 = 4 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: llama.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 11: general.file_type u32 = 2 llama_model_loader: - kv 12: tokenizer.ggml.model str = llama llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,32002] = ["<unk>", "<s>", "</s>", "<0x00>", "<... llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,32002] = [0.000000, 0.000000, 0.000000, 0.0000... llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,32002] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... llama_model_loader: - kv 16: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 17: tokenizer.ggml.eos_token_id u32 = 32000 llama_model_loader: - kv 18: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 19: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 20: general.quantization_version u32 = 2 llama_model_loader: - type f32: 45 tensors llama_model_loader: - type q4_0: 155 tensors llama_model_loader: - type q6_K: 1 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_0 print_info: file size = 606.53 MiB (4.63 BPW) load: control-looking token: 32000 '<|im_end|>' was not control-type; this is probably a bug in the model. its type will be overridden load: special tokens cache size = 5 load: token to piece cache size = 0.1684 MB print_info: arch = llama print_info: vocab_only = 0 print_info: n_ctx_train = 4096 print_info: n_embd = 2048 print_info: n_layer = 22 print_info: n_head = 32 print_info: n_head_kv = 4 print_info: n_rot = 64 print_info: n_swa = 0 print_info: n_swa_pattern = 1 print_info: n_embd_head_k = 64 print_info: n_embd_head_v = 64 print_info: n_gqa = 8 print_info: n_embd_k_gqa = 256 print_info: n_embd_v_gqa = 256 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 = 5632 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 = 0 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: ssm_d_conv = 0 print_info: ssm_d_inner = 0 print_info: ssm_d_state = 0 print_info: ssm_dt_rank = 0 print_info: ssm_dt_b_c_rms = 0 print_info: model type = 1B print_info: model params = 1.10 B print_info: general.name = LLaMA v2 print_info: vocab type = SPM print_info: n_vocab = 32002 print_info: n_merges = 0 print_info: BOS token = 1 '<s>' print_info: EOS token = 32000 '<|im_end|>' print_info: EOT token = 32000 '<|im_end|>' print_info: UNK token = 0 '<unk>' print_info: LF token = 13 '<0x0A>' print_info: EOG token = 32000 '<|im_end|>' print_info: max token length = 48 load_tensors: loading model tensors, this can take a while... (mmap = true) load_tensors: CPU_Mapped model buffer size = 606.53 MiB llama_context: constructing llama_context llama_context: n_seq_max = 2 llama_context: n_ctx = 8192 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 1024 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: freq_base = 10000.0 llama_context: freq_scale = 1 llama_context: CPU output buffer size = 0.26 MiB llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 22, can_shift = 1, padding = 32 llama_kv_cache_unified: CPU KV buffer size = 176.00 MiB llama_kv_cache_unified: KV self size = 176.00 MiB, K (f16): 88.00 MiB, V (f16): 88.00 MiB llama_context: CPU compute buffer size = 544.01 MiB llama_context: graph nodes = 754 llama_context: graph splits = 1 time=2025-06-05T11:50:37.791+02:00 level=INFO source=server.go:630 msg="llama runner started in 0.25 seconds" [GIN] 2025/06/05 - 11:50:37 | 200 | 614.135872ms | 127.0.0.1 | POST "/api/generate" [GIN] 2025/06/05 - 11:50:53 | 200 | 8.678716289s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/06/05 - 11:51:16 | 200 | 15.028160535s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/06/05 - 11:53:54 | 200 | 73.113µs | 127.0.0.1 | GET "/api/version" ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-12 19:15:39 -05:00
GiteaMirror commented 2026-04-12 19:15:41 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 5, 2025):

time=2025-06-05T11:50:37.550+02:00 level=INFO source=runner.go:815 msg="starting go runner"
load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-alderlake.so
time=2025-06-05T11:50:37.553+02:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMA

No GPU enabled backends were found. How did you install ollama?

<!-- gh-comment-id:2943833796 --> @rick-github commented on GitHub (Jun 5, 2025): ``` time=2025-06-05T11:50:37.550+02:00 level=INFO source=runner.go:815 msg="starting go runner" load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-alderlake.so time=2025-06-05T11:50:37.553+02:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMA ``` No GPU enabled backends were found. How did you install ollama?
GiteaMirror commented 2026-04-12 19:15:42 -05:00
Author
Owner
Copy Link

@CastilloDel commented on GitHub (Jun 5, 2025):

time=2025-06-05T11:50:37.550+02:00 level=INFO source=runner.go:815 msg="starting go runner"
load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-alderlake.so
time=2025-06-05T11:50:37.553+02:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMA

No GPU enabled backends were found. How did you install ollama?

Ah you were on the right track. I thought the ollama arch package was enough, but I wasn't aware I needed ollama-cuda and was a bit confused by it still detecting my drivers time=2025-06-05T11:50:28.232+02:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-6b310bc7-c8da-0676-7802-6c2ce1453721 library=cuda variant=v12 compute=8.6 driver=12.8 name="NVIDI.

Thanks!

<!-- gh-comment-id:2944556680 --> @CastilloDel commented on GitHub (Jun 5, 2025): > ``` > time=2025-06-05T11:50:37.550+02:00 level=INFO source=runner.go:815 msg="starting go runner" > load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-alderlake.so > time=2025-06-05T11:50:37.553+02:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMA > ``` > > No GPU enabled backends were found. How did you install ollama? Ah you were on the right track. I thought the `ollama` arch package was enough, but I wasn't aware I needed `ollama-cuda` and was a bit confused by it still detecting my drivers `time=2025-06-05T11:50:28.232+02:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-6b310bc7-c8da-0676-7802-6c2ce1453721 library=cuda variant=v12 compute=8.6 driver=12.8 name="NVIDI`. Thanks!
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#7235
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?