Issue Station
Sign In
Tools hereToolbox

ollama - 💡(How to fix) Fix Windows Ollama 0.22.0 uses CPU only on AMD Ryzen AI 9 HX 470 / Radeon 890M [1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
ollama/ollama#15878•Fetched 2026-04-30 06:18:49
View on GitHub
Comments
0
Participants
1
Timeline
1
Reactions
0
Author
henrytseuk-pixelhenrytseuk-pixel
Participants
henrytseuk-pixelhenrytseuk-pixel
Timeline (top)
labeled ×1

Error Message

time=2026-04-29T14:20:57.089+01:00 level=WARN source=server.go:237 msg="models path not accessible, using default" path=C:\Users\hen_t.ollama\models err="CreateFile C:\Users\hen_t\.ollama\models: The system cannot find the path specified."

Code Example

time=2026-04-29T14:20:56.084+01:00 level=INFO source=app_windows.go:282 msg="starting Ollama" app=C:\Users\hen_t\AppData\Local\Programs\Ollama version=0.22.0 OS=Windows/10.0.26200
time=2026-04-29T14:20:56.088+01:00 level=INFO source=app.go:239 msg="initialized tools registry" tool_count=0
time=2026-04-29T14:20:56.088+01:00 level=INFO source=app.go:254 msg="starting ollama server"
time=2026-04-29T14:20:56.089+01:00 level=INFO source=app.go:285 msg="starting ui server" port=58389
time=2026-04-29T14:20:56.146+01:00 level=INFO source=app_windows.go:268 msg="Created Startup shortcut" shortcut="C:\\Users\\hen_t\\AppData\\Roaming\\Microsoft\\Windows\\Start Menu\\Programs\\Startup\\Ollama.lnk"
time=2026-04-29T14:20:57.089+01:00 level=WARN source=server.go:237 msg="models path not accessible, using default" path=C:\Users\hen_t\.ollama\models err="CreateFile C:\\Users\\hen_t\\.ollama\\models: The system cannot find the path specified."
time=2026-04-29T14:20:59.089+01:00 level=INFO source=updater.go:296 msg="beginning update checker" interval=1h0m0s
time=2026-04-29T14:20:59.089+01:00 level=INFO source=auth.go:62 msg="Failed to load private key: open C:\\Users\\hen_t\\.ollama\\id_ed25519: The system cannot find the path specified."


Couldn't find 'C:\Users\hen_t\.ollama\id_ed25519'. Generating new private key.
Your new public key is: 

ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIJ3wiBzcD+J+7uwUsZJQw04F/337GWStdezGmRYYmwGI

time=2026-04-29T14:20:59.820+01:00 level=INFO source=routes.go:1762 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:0 OLLAMA_DEBUG:INFO OLLAMA_DEBUG_LOG_REQUESTS:false OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 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\\hen_t\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD: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 OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2026-04-29T14:20:59.820+01:00 level=INFO source=routes.go:1764 msg="Ollama cloud disabled: false"
time=2026-04-29T14:20:59.821+01:00 level=INFO source=images.go:517 msg="total blobs: 0"
time=2026-04-29T14:20:59.821+01:00 level=INFO source=images.go:524 msg="total unused blobs removed: 0"
time=2026-04-29T14:20:59.821+01:00 level=INFO source=routes.go:1820 msg="Listening on 127.0.0.1:11434 (version 0.22.0)"
time=2026-04-29T14:20:59.822+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-04-29T14:20:59.834+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58744"
time=2026-04-29T14:21:27.391+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58754"
time=2026-04-29T14:21:34.953+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58766"
time=2026-04-29T14:21:42.792+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled.  To enable, set OLLAMA_VULKAN=1"
time=2026-04-29T14:21:42.806+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58772"
time=2026-04-29T14:21:43.062+01:00 level=INFO source=types.go:60 msg="inference compute" id=cpu library=cpu compute="" name=cpu description=cpu libdirs=ollama driver="" pci_id="" type="" total="111.6 GiB" available="83.6 GiB"
time=2026-04-29T14:21:43.062+01:00 level=INFO source=routes.go:1870 msg="vram-based default context" total_vram="0 B" default_num_ctx=4096
[GIN] 2026/04/29 - 14:21:43 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:21:43 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:21:43 | 200 |      1.6079ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/04/29 - 14:21:43 | 200 |       529.3µs |       127.0.0.1 | GET      "/api/status"
[GIN] 2026/04/29 - 14:21:52 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:21:54 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:22:04 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:22:09 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:22:11 | 200 |            0s |       127.0.0.1 | HEAD     "/"
time=2026-04-29T14:22:12.304+01:00 level=INFO source=download.go:179 msg="downloading 74701a8c35f6 in 14 100 MB part(s)"
time=2026-04-29T14:22:39.213+01:00 level=INFO source=download.go:179 msg="downloading 966de95ca8a6 in 1 1.4 KB part(s)"
time=2026-04-29T14:22:40.514+01:00 level=INFO source=download.go:179 msg="downloading fcc5a6bec9da in 1 7.7 KB part(s)"
time=2026-04-29T14:22:41.815+01:00 level=INFO source=download.go:179 msg="downloading a70ff7e570d9 in 1 6.0 KB part(s)"
time=2026-04-29T14:22:43.128+01:00 level=INFO source=download.go:179 msg="downloading 4f659a1e86d7 in 1 485 B part(s)"
[GIN] 2026/04/29 - 14:22:46 | 200 |   35.2083808s |       127.0.0.1 | POST     "/api/pull"
[GIN] 2026/04/29 - 14:23:03 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:23:03 | 200 |    123.8675ms |       127.0.0.1 | POST     "/api/show"
time=2026-04-29T14:23:03.396+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-04-29T14:23:03.396+01:00 level=INFO source=cpu_windows.go:164 msg="efficiency cores detected" maxEfficiencyClass=1
time=2026-04-29T14:23:03.396+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=12 efficiency=8 threads=24
llama_model_loader: loaded meta data with 30 key-value pairs and 147 tensors from C:\Users\hen_t\.ollama\models\blobs\sha256-74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45 (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              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Llama 3.2 1B Instruct
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                           general.basename str              = Llama-3.2
llama_model_loader: - kv   5:                         general.size_label str              = 1B
llama_model_loader: - kv   6:                               general.tags arr[str,6]       = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv   7:                          general.languages arr[str,8]       = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv   8:                          llama.block_count u32              = 16
llama_model_loader: - kv   9:                       llama.context_length u32              = 131072
llama_model_loader: - kv  10:                     llama.embedding_length u32              = 2048
llama_model_loader: - kv  11:                  llama.feed_forward_length u32              = 8192
llama_model_loader: - kv  12:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  13:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  14:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv  15:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  16:                 llama.attention.key_length u32              = 64
llama_model_loader: - kv  17:               llama.attention.value_length u32              = 64
llama_model_loader: - kv  18:                          general.file_type u32              = 7
llama_model_loader: - kv  19:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  20:                 llama.rope.dimension_count u32              = 64
llama_model_loader: - kv  21:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  22:                         tokenizer.ggml.pre str              = llama-bpe
llama_model_loader: - kv  23:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  24:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  25:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  26:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  27:                tokenizer.ggml.eos_token_id u32              = 128009
llama_model_loader: - kv  28:                    tokenizer.chat_template str              = {{- bos_token }}\n{%- if custom_tools ...
llama_model_loader: - kv  29:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   34 tensors
llama_model_loader: - type q8_0:  113 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q8_0
print_info: file size   = 1.22 GiB (8.50 BPW) 
load: printing all EOG tokens:
load:   - 128001 ('<|end_of_text|>')
load:   - 128008 ('<|eom_id|>')
load:   - 128009 ('<|eot_id|>')
load: special tokens cache size = 256
load: token to piece cache size = 0.7999 MB
print_info: arch             = llama
print_info: vocab_only       = 1
print_info: no_alloc         = 0
print_info: model type       = ?B
print_info: model params     = 1.24 B
print_info: general.name     = Llama 3.2 1B Instruct
print_info: vocab type       = BPE
print_info: n_vocab          = 128256
print_info: n_merges         = 280147
print_info: BOS token        = 128000 '<|begin_of_text|>'
print_info: EOS token        = 128009 '<|eot_id|>'
print_info: EOT token        = 128001 '<|end_of_text|>'
print_info: EOM token        = 128008 '<|eom_id|>'
print_info: LF token         = 198 'Ċ'
print_info: EOG token        = 128001 '<|end_of_text|>'
print_info: EOG token        = 128008 '<|eom_id|>'
print_info: EOG token        = 128009 '<|eot_id|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-04-29T14:23:03.886+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\hen_t\\.ollama\\models\\blobs\\sha256-74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45 --port 63009"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=sched.go:484 msg="system memory" total="111.6 GiB" free="84.5 GiB" free_swap="106.1 GiB"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=server.go:511 msg="loading model" "model layers"=17 requested=-1
time=2026-04-29T14:23:03.893+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="1.2 GiB"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="128.0 MiB"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=device.go:272 msg="total memory" size="1.3 GiB"
time=2026-04-29T14:23:03.933+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\hen_t\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
time=2026-04-29T14:23:03.947+01:00 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.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(clang)
time=2026-04-29T14:23:03.947+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:63009"
time=2026-04-29T14:23:03.952+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-04-29T14:23:03.952+01:00 level=INFO source=server.go:1364 msg="waiting for llama runner to start responding"
time=2026-04-29T14:23:03.952+01:00 level=INFO source=server.go:1398 msg="waiting for server to become available" status="llm server loading model"
llama_model_loader: loaded meta data with 30 key-value pairs and 147 tensors from C:\Users\hen_t\.ollama\models\blobs\sha256-74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45 (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              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Llama 3.2 1B Instruct
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                           general.basename str              = Llama-3.2
llama_model_loader: - kv   5:                         general.size_label str              = 1B
llama_model_loader: - kv   6:                               general.tags arr[str,6]       = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv   7:                          general.languages arr[str,8]       = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv   8:                          llama.block_count u32              = 16
llama_model_loader: - kv   9:                       llama.context_length u32              = 131072
llama_model_loader: - kv  10:                     llama.embedding_length u32              = 2048
llama_model_loader: - kv  11:                  llama.feed_forward_length u32              = 8192
llama_model_loader: - kv  12:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  13:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  14:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv  15:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  16:                 llama.attention.key_length u32              = 64
llama_model_loader: - kv  17:               llama.attention.value_length u32              = 64
llama_model_loader: - kv  18:                          general.file_type u32              = 7
llama_model_loader: - kv  19:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  20:                 llama.rope.dimension_count u32              = 64
llama_model_loader: - kv  21:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  22:                         tokenizer.ggml.pre str              = llama-bpe
llama_model_loader: - kv  23:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  24:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  25:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  26:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  27:                tokenizer.ggml.eos_token_id u32              = 128009
llama_model_loader: - kv  28:                    tokenizer.chat_template str              = {{- bos_token }}\n{%- if custom_tools ...
llama_model_loader: - kv  29:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   34 tensors
llama_model_loader: - type q8_0:  113 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q8_0
print_info: file size   = 1.22 GiB (8.50 BPW) 
load: printing all EOG tokens:
load:   - 128001 ('<|end_of_text|>')
load:   - 128008 ('<|eom_id|>')
load:   - 128009 ('<|eot_id|>')
load: special tokens cache size = 256
load: token to piece cache size = 0.7999 MB
print_info: arch             = llama
print_info: vocab_only       = 0
print_info: no_alloc         = 0
print_info: n_ctx_train      = 131072
print_info: n_embd           = 2048
print_info: n_embd_inp       = 2048
print_info: n_layer          = 16
print_info: n_head           = 32
print_info: n_head_kv        = 8
print_info: n_rot            = 64
print_info: n_swa            = 0
print_info: is_swa_any       = 0
print_info: n_embd_head_k    = 64
print_info: n_embd_head_v    = 64
print_info: n_gqa            = 4
print_info: n_embd_k_gqa     = 512
print_info: n_embd_v_gqa     = 512
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: n_expert_groups  = 0
print_info: n_group_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  = 500000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned   = unknown
print_info: model type       = 1B
print_info: model params     = 1.24 B
print_info: general.name     = Llama 3.2 1B Instruct
print_info: vocab type       = BPE
print_info: n_vocab          = 128256
print_info: n_merges         = 280147
print_info: BOS token        = 128000 '<|begin_of_text|>'
print_info: EOS token        = 128009 '<|eot_id|>'
print_info: EOT token        = 128001 '<|end_of_text|>'
print_info: EOM token        = 128008 '<|eom_id|>'
print_info: LF token         = 198 'Ċ'
print_info: EOG token        = 128001 '<|end_of_text|>'
print_info: EOG token        = 128008 '<|eom_id|>'
print_info: EOG token        = 128009 '<|eot_id|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors:          CPU model buffer size =  1252.41 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 4096
llama_context: n_ctx_seq     = 4096
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = auto
llama_context: kv_unified    = false
llama_context: freq_base     = 500000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     0.50 MiB
llama_kv_cache:        CPU KV buffer size =   128.00 MiB
llama_kv_cache: size =  128.00 MiB (  4096 cells,  16 layers,  1/1 seqs), K (f16):   64.00 MiB, V (f16):   64.00 MiB
llama_context: Flash Attention was auto, set to enabled
llama_context:        CPU compute buffer size =   258.50 MiB
llama_context: graph nodes  = 503
llama_context: graph splits = 1
time=2026-04-29T14:23:04.454+01:00 level=INFO source=server.go:1402 msg="llama runner started in 0.56 seconds"
time=2026-04-29T14:23:04.455+01:00 level=INFO source=sched.go:561 msg="loaded runners" count=1
time=2026-04-29T14:23:04.455+01:00 level=INFO source=server.go:1364 msg="waiting for llama runner to start responding"
time=2026-04-29T14:23:04.455+01:00 level=INFO source=server.go:1402 msg="llama runner started in 0.56 seconds"
[GIN] 2026/04/29 - 14:23:08 | 200 |    5.6175667s |       127.0.0.1 | POST     "/api/generate"
[GIN] 2026/04/29 - 14:23:48 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:23:48 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:24:03 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:24:03 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:26:06 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:26:07 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:26:07 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:28:21 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:28:22 | 200 |       506.4µs |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:28:22 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:29:02 | 200 |      1.6282ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
RAW_BUFFERClick to expand / collapse

What is the issue?

I am raising this after a clean reinstall and controlled test.

Hardware:

  • Minisforum AI X1 Pro-470
  • AMD Ryzen AI 9 HX 470
  • Integrated AMD Radeon 890M Graphics
  • 128GB RAM
  • Windows 11 Pro
  • UMA frame buffer previously tested at 16GB

GPU / driver:

  • AMD Radeon(TM) 890M Graphics
  • Driver version: 32.0.23033.1002
  • Driver date: 09/03/2026

Ollama:

  • Official Windows install
  • Ollama version: 0.22.0
  • Install path: C:\Users\hen_t\AppData\Local\Programs\Ollama\ollama.exe
  • API works on http://127.0.0.1:11434

Clean-up before reinstall:

  • Removed previous AMD AI Bundle Ollama
  • Removed WSL Ollama
  • Removed Docker Ollama image/container/volume
  • Confirmed no previous Ollama process, command, port 11434 listener, PATH entry, model store, or startup shortcut before reinstall

Test:

  1. Installed official Windows Ollama
  2. Ran: ollama pull llama3.2:1b
  3. Ran: ollama run llama3.2:1b
  4. Ran: ollama ps

Result:

  • Model runs successfully
  • ollama ps reports: 100% CPU
  • No GPU offload observed, even with small llama3.2:1b model

Observed process:

  • ollama.exe serve
  • ollama.exe
  • ollama.exe runner --model ...

Question: Does official Windows Ollama currently support GPU acceleration on AMD Ryzen AI 9 HX 470 / Radeon 890M / gfx1150? If not, is support planned? If yes, what diagnostic logs or environment checks would help determine why Ollama falls back to CPU?

I can provide server.log, ollama ps output, GPU driver details

Relevant log output

time=2026-04-29T14:20:56.084+01:00 level=INFO source=app_windows.go:282 msg="starting Ollama" app=C:\Users\hen_t\AppData\Local\Programs\Ollama version=0.22.0 OS=Windows/10.0.26200
time=2026-04-29T14:20:56.088+01:00 level=INFO source=app.go:239 msg="initialized tools registry" tool_count=0
time=2026-04-29T14:20:56.088+01:00 level=INFO source=app.go:254 msg="starting ollama server"
time=2026-04-29T14:20:56.089+01:00 level=INFO source=app.go:285 msg="starting ui server" port=58389
time=2026-04-29T14:20:56.146+01:00 level=INFO source=app_windows.go:268 msg="Created Startup shortcut" shortcut="C:\\Users\\hen_t\\AppData\\Roaming\\Microsoft\\Windows\\Start Menu\\Programs\\Startup\\Ollama.lnk"
time=2026-04-29T14:20:57.089+01:00 level=WARN source=server.go:237 msg="models path not accessible, using default" path=C:\Users\hen_t\.ollama\models err="CreateFile C:\\Users\\hen_t\\.ollama\\models: The system cannot find the path specified."
time=2026-04-29T14:20:59.089+01:00 level=INFO source=updater.go:296 msg="beginning update checker" interval=1h0m0s
time=2026-04-29T14:20:59.089+01:00 level=INFO source=auth.go:62 msg="Failed to load private key: open C:\\Users\\hen_t\\.ollama\\id_ed25519: The system cannot find the path specified."


Couldn't find 'C:\Users\hen_t\.ollama\id_ed25519'. Generating new private key.
Your new public key is: 

ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIJ3wiBzcD+J+7uwUsZJQw04F/337GWStdezGmRYYmwGI

time=2026-04-29T14:20:59.820+01:00 level=INFO source=routes.go:1762 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:0 OLLAMA_DEBUG:INFO OLLAMA_DEBUG_LOG_REQUESTS:false OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 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\\hen_t\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD: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 OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2026-04-29T14:20:59.820+01:00 level=INFO source=routes.go:1764 msg="Ollama cloud disabled: false"
time=2026-04-29T14:20:59.821+01:00 level=INFO source=images.go:517 msg="total blobs: 0"
time=2026-04-29T14:20:59.821+01:00 level=INFO source=images.go:524 msg="total unused blobs removed: 0"
time=2026-04-29T14:20:59.821+01:00 level=INFO source=routes.go:1820 msg="Listening on 127.0.0.1:11434 (version 0.22.0)"
time=2026-04-29T14:20:59.822+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-04-29T14:20:59.834+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58744"
time=2026-04-29T14:21:27.391+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58754"
time=2026-04-29T14:21:34.953+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58766"
time=2026-04-29T14:21:42.792+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled.  To enable, set OLLAMA_VULKAN=1"
time=2026-04-29T14:21:42.806+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 58772"
time=2026-04-29T14:21:43.062+01:00 level=INFO source=types.go:60 msg="inference compute" id=cpu library=cpu compute="" name=cpu description=cpu libdirs=ollama driver="" pci_id="" type="" total="111.6 GiB" available="83.6 GiB"
time=2026-04-29T14:21:43.062+01:00 level=INFO source=routes.go:1870 msg="vram-based default context" total_vram="0 B" default_num_ctx=4096
[GIN] 2026/04/29 - 14:21:43 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:21:43 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:21:43 | 200 |      1.6079ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/04/29 - 14:21:43 | 200 |       529.3µs |       127.0.0.1 | GET      "/api/status"
[GIN] 2026/04/29 - 14:21:52 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:21:54 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:22:04 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:22:09 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:22:11 | 200 |            0s |       127.0.0.1 | HEAD     "/"
time=2026-04-29T14:22:12.304+01:00 level=INFO source=download.go:179 msg="downloading 74701a8c35f6 in 14 100 MB part(s)"
time=2026-04-29T14:22:39.213+01:00 level=INFO source=download.go:179 msg="downloading 966de95ca8a6 in 1 1.4 KB part(s)"
time=2026-04-29T14:22:40.514+01:00 level=INFO source=download.go:179 msg="downloading fcc5a6bec9da in 1 7.7 KB part(s)"
time=2026-04-29T14:22:41.815+01:00 level=INFO source=download.go:179 msg="downloading a70ff7e570d9 in 1 6.0 KB part(s)"
time=2026-04-29T14:22:43.128+01:00 level=INFO source=download.go:179 msg="downloading 4f659a1e86d7 in 1 485 B part(s)"
[GIN] 2026/04/29 - 14:22:46 | 200 |   35.2083808s |       127.0.0.1 | POST     "/api/pull"
[GIN] 2026/04/29 - 14:23:03 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:23:03 | 200 |    123.8675ms |       127.0.0.1 | POST     "/api/show"
time=2026-04-29T14:23:03.396+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-04-29T14:23:03.396+01:00 level=INFO source=cpu_windows.go:164 msg="efficiency cores detected" maxEfficiencyClass=1
time=2026-04-29T14:23:03.396+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=12 efficiency=8 threads=24
llama_model_loader: loaded meta data with 30 key-value pairs and 147 tensors from C:\Users\hen_t\.ollama\models\blobs\sha256-74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45 (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              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Llama 3.2 1B Instruct
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                           general.basename str              = Llama-3.2
llama_model_loader: - kv   5:                         general.size_label str              = 1B
llama_model_loader: - kv   6:                               general.tags arr[str,6]       = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv   7:                          general.languages arr[str,8]       = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv   8:                          llama.block_count u32              = 16
llama_model_loader: - kv   9:                       llama.context_length u32              = 131072
llama_model_loader: - kv  10:                     llama.embedding_length u32              = 2048
llama_model_loader: - kv  11:                  llama.feed_forward_length u32              = 8192
llama_model_loader: - kv  12:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  13:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  14:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv  15:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  16:                 llama.attention.key_length u32              = 64
llama_model_loader: - kv  17:               llama.attention.value_length u32              = 64
llama_model_loader: - kv  18:                          general.file_type u32              = 7
llama_model_loader: - kv  19:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  20:                 llama.rope.dimension_count u32              = 64
llama_model_loader: - kv  21:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  22:                         tokenizer.ggml.pre str              = llama-bpe
llama_model_loader: - kv  23:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  24:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  25:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  26:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  27:                tokenizer.ggml.eos_token_id u32              = 128009
llama_model_loader: - kv  28:                    tokenizer.chat_template str              = {{- bos_token }}\n{%- if custom_tools ...
llama_model_loader: - kv  29:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   34 tensors
llama_model_loader: - type q8_0:  113 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q8_0
print_info: file size   = 1.22 GiB (8.50 BPW) 
load: printing all EOG tokens:
load:   - 128001 ('<|end_of_text|>')
load:   - 128008 ('<|eom_id|>')
load:   - 128009 ('<|eot_id|>')
load: special tokens cache size = 256
load: token to piece cache size = 0.7999 MB
print_info: arch             = llama
print_info: vocab_only       = 1
print_info: no_alloc         = 0
print_info: model type       = ?B
print_info: model params     = 1.24 B
print_info: general.name     = Llama 3.2 1B Instruct
print_info: vocab type       = BPE
print_info: n_vocab          = 128256
print_info: n_merges         = 280147
print_info: BOS token        = 128000 '<|begin_of_text|>'
print_info: EOS token        = 128009 '<|eot_id|>'
print_info: EOT token        = 128001 '<|end_of_text|>'
print_info: EOM token        = 128008 '<|eom_id|>'
print_info: LF token         = 198 'Ċ'
print_info: EOG token        = 128001 '<|end_of_text|>'
print_info: EOG token        = 128008 '<|eom_id|>'
print_info: EOG token        = 128009 '<|eot_id|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-04-29T14:23:03.886+01:00 level=INFO source=server.go:444 msg="starting runner" cmd="C:\\Users\\hen_t\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\hen_t\\.ollama\\models\\blobs\\sha256-74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45 --port 63009"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=sched.go:484 msg="system memory" total="111.6 GiB" free="84.5 GiB" free_swap="106.1 GiB"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=server.go:511 msg="loading model" "model layers"=17 requested=-1
time=2026-04-29T14:23:03.893+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="1.2 GiB"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="128.0 MiB"
time=2026-04-29T14:23:03.893+01:00 level=INFO source=device.go:272 msg="total memory" size="1.3 GiB"
time=2026-04-29T14:23:03.933+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\hen_t\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
time=2026-04-29T14:23:03.947+01:00 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.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(clang)
time=2026-04-29T14:23:03.947+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:63009"
time=2026-04-29T14:23:03.952+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-04-29T14:23:03.952+01:00 level=INFO source=server.go:1364 msg="waiting for llama runner to start responding"
time=2026-04-29T14:23:03.952+01:00 level=INFO source=server.go:1398 msg="waiting for server to become available" status="llm server loading model"
llama_model_loader: loaded meta data with 30 key-value pairs and 147 tensors from C:\Users\hen_t\.ollama\models\blobs\sha256-74701a8c35f6c8d9a4b91f3f3497643001d63e0c7a84e085bed452548fa88d45 (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              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Llama 3.2 1B Instruct
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                           general.basename str              = Llama-3.2
llama_model_loader: - kv   5:                         general.size_label str              = 1B
llama_model_loader: - kv   6:                               general.tags arr[str,6]       = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv   7:                          general.languages arr[str,8]       = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv   8:                          llama.block_count u32              = 16
llama_model_loader: - kv   9:                       llama.context_length u32              = 131072
llama_model_loader: - kv  10:                     llama.embedding_length u32              = 2048
llama_model_loader: - kv  11:                  llama.feed_forward_length u32              = 8192
llama_model_loader: - kv  12:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  13:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  14:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv  15:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  16:                 llama.attention.key_length u32              = 64
llama_model_loader: - kv  17:               llama.attention.value_length u32              = 64
llama_model_loader: - kv  18:                          general.file_type u32              = 7
llama_model_loader: - kv  19:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  20:                 llama.rope.dimension_count u32              = 64
llama_model_loader: - kv  21:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  22:                         tokenizer.ggml.pre str              = llama-bpe
llama_model_loader: - kv  23:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  24:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  25:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  26:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  27:                tokenizer.ggml.eos_token_id u32              = 128009
llama_model_loader: - kv  28:                    tokenizer.chat_template str              = {{- bos_token }}\n{%- if custom_tools ...
llama_model_loader: - kv  29:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   34 tensors
llama_model_loader: - type q8_0:  113 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q8_0
print_info: file size   = 1.22 GiB (8.50 BPW) 
load: printing all EOG tokens:
load:   - 128001 ('<|end_of_text|>')
load:   - 128008 ('<|eom_id|>')
load:   - 128009 ('<|eot_id|>')
load: special tokens cache size = 256
load: token to piece cache size = 0.7999 MB
print_info: arch             = llama
print_info: vocab_only       = 0
print_info: no_alloc         = 0
print_info: n_ctx_train      = 131072
print_info: n_embd           = 2048
print_info: n_embd_inp       = 2048
print_info: n_layer          = 16
print_info: n_head           = 32
print_info: n_head_kv        = 8
print_info: n_rot            = 64
print_info: n_swa            = 0
print_info: is_swa_any       = 0
print_info: n_embd_head_k    = 64
print_info: n_embd_head_v    = 64
print_info: n_gqa            = 4
print_info: n_embd_k_gqa     = 512
print_info: n_embd_v_gqa     = 512
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: n_expert_groups  = 0
print_info: n_group_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  = 500000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned   = unknown
print_info: model type       = 1B
print_info: model params     = 1.24 B
print_info: general.name     = Llama 3.2 1B Instruct
print_info: vocab type       = BPE
print_info: n_vocab          = 128256
print_info: n_merges         = 280147
print_info: BOS token        = 128000 '<|begin_of_text|>'
print_info: EOS token        = 128009 '<|eot_id|>'
print_info: EOT token        = 128001 '<|end_of_text|>'
print_info: EOM token        = 128008 '<|eom_id|>'
print_info: LF token         = 198 'Ċ'
print_info: EOG token        = 128001 '<|end_of_text|>'
print_info: EOG token        = 128008 '<|eom_id|>'
print_info: EOG token        = 128009 '<|eot_id|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors:          CPU model buffer size =  1252.41 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 4096
llama_context: n_ctx_seq     = 4096
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = auto
llama_context: kv_unified    = false
llama_context: freq_base     = 500000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     0.50 MiB
llama_kv_cache:        CPU KV buffer size =   128.00 MiB
llama_kv_cache: size =  128.00 MiB (  4096 cells,  16 layers,  1/1 seqs), K (f16):   64.00 MiB, V (f16):   64.00 MiB
llama_context: Flash Attention was auto, set to enabled
llama_context:        CPU compute buffer size =   258.50 MiB
llama_context: graph nodes  = 503
llama_context: graph splits = 1
time=2026-04-29T14:23:04.454+01:00 level=INFO source=server.go:1402 msg="llama runner started in 0.56 seconds"
time=2026-04-29T14:23:04.455+01:00 level=INFO source=sched.go:561 msg="loaded runners" count=1
time=2026-04-29T14:23:04.455+01:00 level=INFO source=server.go:1364 msg="waiting for llama runner to start responding"
time=2026-04-29T14:23:04.455+01:00 level=INFO source=server.go:1402 msg="llama runner started in 0.56 seconds"
[GIN] 2026/04/29 - 14:23:08 | 200 |    5.6175667s |       127.0.0.1 | POST     "/api/generate"
[GIN] 2026/04/29 - 14:23:48 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:23:48 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:24:03 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:24:03 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:26:06 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:26:07 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:26:07 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:28:21 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:28:22 | 200 |       506.4µs |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:28:22 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:29:02 | 200 |      1.6282ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/04/29 - 14:29:02 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"

OS

Windows

GPU

AMD

CPU

AMD

Ollama version

0.22.0

extent analysis

TL;DR

The issue is likely due to the lack of GPU acceleration support for AMD Ryzen AI 9 HX 470 / Radeon 890M in the current Ollama version, causing the model to run on CPU.

Guidance

  1. Check Ollama documentation: Verify if the current version of Ollama supports GPU acceleration on AMD Ryzen AI 9 HX 470 / Radeon 890M.
  2. Enable Vulkan support: Set the environment variable OLLAMA_VULKAN to 1 to enable experimental Vulkan support, which might allow GPU acceleration.
  3. Monitor system resources: Use tools like Task Manager to monitor CPU and GPU usage to confirm if the model is running on CPU or GPU.
  4. Check for updates: Look for updates to Ollama that may include support for AMD GPUs.

Example

To enable Vulkan support, set the environment variable OLLAMA_VULKAN to 1 before running Ollama:

set OLLAMA_VULKAN=1
ollama run llama3.2:1b

Notes

The provided log output indicates that the model is running on CPU, and there is no evidence of GPU acceleration. The OLLAMA_VULKAN environment variable is set to false by default, which might be the cause of the issue.

Recommendation

Apply workaround: Enable Vulkan support by setting OLLAMA_VULKAN to 1 to potentially allow GPU acceleration.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#ssr#docker error#permission error#memory optimization

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

ollama - 💡(How to fix) Fix Windows Ollama 0.22.0 uses CPU only on AMD Ryzen AI 9 HX 470 / Radeon 890M [1 participants]