github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10218] Image recognition doesn't work with models downloaded from another site #53217

New Issue
Closed
opened 2026-04-29 02:23:44 -05:00 by GiteaMirror · 9 comments
GiteaMirror commented 2026-04-29 02:23:44 -05:00
Owner
Copy Link

Originally created by @yukkuriTV on GitHub (Apr 10, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10218

What is the issue?

To troubleshoot, I tried using gemma-3-4b-it-Q8_0 that I downloaded from another site, and although it recognizes and uses text fine, it doesn't generate an answer when I attach an image.

Even if you provide an answer, nothing is generated.

Relevant log output

2025-04-10 23:51:34.615 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50141 - "POST /api/v1/chats/e001df23-806c-4e32-b5c2-c71ccd8d9a8d HTTP/1.1" 200 - {}
2025-04-10 23:51:34.632 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50141 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}
2025-04-10 23:51:39.754 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50150 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}
2025-04-10 23:51:44.792 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50152 - "GET /api/v1/chats/e001df23-806c-4e32-b5c2-c71ccd8d9a8d HTTP/1.1" 200 - {}
2025-04-10 23:51:44.801 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50152 - "GET /api/v1/chats/all/tags HTTP/1.1" 200 - {}
2025-04-10 23:52:08.090 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /api/v1/chats/7e3f597a-f490-497c-ab9f-83cc5e72a4c5 HTTP/1.1" 200 - {}
2025-04-10 23:52:14.218 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50157 - "GET /static/favicon.png HTTP/1.1" 304 - {}
2025-04-10 23:52:14.232 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50161 - "GET /api/v1/chats/087c93bf-00a0-41f5-830f-8a563f521156 HTTP/1.1" 200 - {}
2025-04-10 23:52:14.234 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50162 - "GET /api/v1/chats/72d2437f-43b2-49c6-8c6a-408710f3dcb3 HTTP/1.1" 200 - {}
2025-04-10 23:52:14.234 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50163 - "GET /api/v1/chats/82efb2ba-9705-4905-b9ca-d7f1058e8d3f HTTP/1.1" 200 - {}
2025-04-10 23:52:14.274 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /api/v1/chats/7e3f597a-f490-497c-ab9f-83cc5e72a4c5/tags HTTP/1.1" 200 - {}
2025-04-10 23:52:14.283 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /api/v1/users/user/settings HTTP/1.1" 200 - {}
2025-04-10 23:52:14.567 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /ollama/api/version HTTP/1.1" 200 - {}
2025-04-10 23:52:21.128 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50171 - "GET /api/v1/users/user/settings HTTP/1.1" 200 - {}
2025-04-10 23:52:21.298 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50169 - "GET /ollama/api/version HTTP/1.1" 200 - {}
2025-04-10 23:52:54.623 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/chats/new HTTP/1.1" 200 - {}
2025-04-10 23:52:58.774 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}
2025-04-10 23:52:58.821 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/chats/4a8eda46-61ae-4a95-a89f-09be2513a31a HTTP/1.1" 200 - {}
2025-04-10 23:52:58.901 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}
Batches: 100%|██████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 102.54it/s]
2025-04-10 23:52:58.926 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/memories/query HTTP/1.1" 200 - {}
2025-04-10 23:52:59.290 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/chat/completions HTTP/1.1" 200 - {}
2025-04-10 23:52:59.297 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/chat/completed HTTP/1.1" 200 - {}
2025-04-10 23:52:59.308 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}
2025-04-10 23:52:59.330 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/chats/4a8eda46-61ae-4a95-a89f-09be2513a31a HTTP/1.1" 200 - {}
2025-04-10 23:52:59.342 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}
2025-04-10 23:53:04.100 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {}

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.6.5

Originally created by @yukkuriTV on GitHub (Apr 10, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10218 ### What is the issue? To troubleshoot, I tried using gemma-3-4b-it-Q8_0 that I downloaded from another site, and although it recognizes and uses text fine, it doesn't generate an answer when I attach an image. Even if you provide an answer, nothing is generated. ### Relevant log output ```shell 2025-04-10 23:51:34.615 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50141 - "POST /api/v1/chats/e001df23-806c-4e32-b5c2-c71ccd8d9a8d HTTP/1.1" 200 - {} 2025-04-10 23:51:34.632 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50141 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} 2025-04-10 23:51:39.754 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50150 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} 2025-04-10 23:51:44.792 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50152 - "GET /api/v1/chats/e001df23-806c-4e32-b5c2-c71ccd8d9a8d HTTP/1.1" 200 - {} 2025-04-10 23:51:44.801 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50152 - "GET /api/v1/chats/all/tags HTTP/1.1" 200 - {} 2025-04-10 23:52:08.090 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /api/v1/chats/7e3f597a-f490-497c-ab9f-83cc5e72a4c5 HTTP/1.1" 200 - {} 2025-04-10 23:52:14.218 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50157 - "GET /static/favicon.png HTTP/1.1" 304 - {} 2025-04-10 23:52:14.232 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50161 - "GET /api/v1/chats/087c93bf-00a0-41f5-830f-8a563f521156 HTTP/1.1" 200 - {} 2025-04-10 23:52:14.234 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50162 - "GET /api/v1/chats/72d2437f-43b2-49c6-8c6a-408710f3dcb3 HTTP/1.1" 200 - {} 2025-04-10 23:52:14.234 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50163 - "GET /api/v1/chats/82efb2ba-9705-4905-b9ca-d7f1058e8d3f HTTP/1.1" 200 - {} 2025-04-10 23:52:14.274 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /api/v1/chats/7e3f597a-f490-497c-ab9f-83cc5e72a4c5/tags HTTP/1.1" 200 - {} 2025-04-10 23:52:14.283 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /api/v1/users/user/settings HTTP/1.1" 200 - {} 2025-04-10 23:52:14.567 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50153 - "GET /ollama/api/version HTTP/1.1" 200 - {} 2025-04-10 23:52:21.128 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50171 - "GET /api/v1/users/user/settings HTTP/1.1" 200 - {} 2025-04-10 23:52:21.298 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50169 - "GET /ollama/api/version HTTP/1.1" 200 - {} 2025-04-10 23:52:54.623 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/chats/new HTTP/1.1" 200 - {} 2025-04-10 23:52:58.774 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} 2025-04-10 23:52:58.821 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/chats/4a8eda46-61ae-4a95-a89f-09be2513a31a HTTP/1.1" 200 - {} 2025-04-10 23:52:58.901 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} Batches: 100%|██████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 102.54it/s] 2025-04-10 23:52:58.926 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/memories/query HTTP/1.1" 200 - {} 2025-04-10 23:52:59.290 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/chat/completions HTTP/1.1" 200 - {} 2025-04-10 23:52:59.297 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/chat/completed HTTP/1.1" 200 - {} 2025-04-10 23:52:59.308 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} 2025-04-10 23:52:59.330 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "POST /api/v1/chats/4a8eda46-61ae-4a95-a89f-09be2513a31a HTTP/1.1" 200 - {} 2025-04-10 23:52:59.342 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} 2025-04-10 23:53:04.100 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:50173 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 - {} ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.6.5
GiteaMirror added the bug label 2026-04-29 02:23:44 -05:00
GiteaMirror commented 2026-04-29 02:23:46 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 10, 2025):

This log is not from ollama. If the model is from a different site, it may not have preserved the vision component of the model when it was quantized.

<!-- gh-comment-id:2794144223 --> @rick-github commented on GitHub (Apr 10, 2025): This log is not from ollama. If the model is from a different site, it may not have preserved the vision component of the model when it was quantized.
GiteaMirror commented 2026-04-29 02:23:46 -05:00
Author
Owner
Copy Link

@yukkuriTV commented on GitHub (Apr 10, 2025):

Sorry, here is the correct one.
2025/04/10 23:41:22 routes.go:1231: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION: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\yukkuriTV\.ollama\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-04-10T23:41:22.262+09:00 level=INFO source=images.go:458 msg="total blobs: 21"
time=2025-04-10T23:41:22.264+09:00 level=INFO source=images.go:465 msg="total unused blobs removed: 0"
time=2025-04-10T23:41:22.267+09:00 level=INFO source=routes.go:1298 msg="Listening on 127.0.0.1:11434 (version 0.6.5)"
time=2025-04-10T23:41:22.269+09:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-04-10T23:41:22.269+09:00 level=INFO source=gpu_windows.go:167 msg=packages count=1
time=2025-04-10T23:41:22.269+09:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12
time=2025-04-10T23:41:22.535+09:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-23f2619b-2453-9ed8-f622-383f1a5428fb library=cuda compute=7.5 driver=12.6 name="NVIDIA GeForce GTX 1650" overhead="445.1 MiB"
time=2025-04-10T23:41:22.538+09:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-23f2619b-2453-9ed8-f622-383f1a5428fb library=cuda variant=v12 compute=7.5 driver=12.6 name="NVIDIA GeForce GTX 1650" total="4.0 GiB" available="3.2 GiB"
[GIN] 2025/04/10 - 23:43:48 | 200 | 77.7487ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:43:49 | 200 | 2.5884ms | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:47:31 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:48:16 | 200 | 4.151ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:48:16 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:48:31 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:48:37 | 200 | 7.7663ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:48:54 | 200 | 6.9112ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:48:55 | 200 | 2.0605ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:04 | 200 | 2.5982ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:05 | 200 | 12.318ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:13 | 200 | 1.5331ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:13 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:49:19 | 200 | 4.7153ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:29 | 200 | 1.5366ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:39 | 200 | 1.5306ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:40 | 200 | 1.531ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:47 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:49:53 | 200 | 1.5666ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:49:59 | 200 | 2.5731ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:50:00 | 200 | 2.0272ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:50:01 | 200 | 1.567ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:50:03 | 200 | 2.1286ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:50:07 | 200 | 0s | 127.0.0.1 | GET "/api/version"
time=2025-04-10T23:50:17.896+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:17.896+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:17.897+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:17.897+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:17.908+09:00 level=INFO source=server.go:105 msg="system memory" total="11.8 GiB" free="5.3 GiB" free_swap="35.1 GiB"
time=2025-04-10T23:50:17.908+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:17.908+09:00 level=INFO source=server.go:138 msg=offload library=cuda layers.requested=-1 layers.model=35 layers.offload=15 layers.split="" memory.available="[3.1 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.6 GiB" memory.required.partial="3.1 GiB" memory.required.kv="214.0 MiB" memory.required.allocations="[3.1 GiB]" memory.weights.total="3.8 GiB" memory.weights.repeating="3.2 GiB" memory.weights.nonrepeating="680.2 MiB" memory.graph.full="517.1 MiB" memory.graph.partial="1.0 GiB"
time=2025-04-10T23:50:17.975+09:00 level=WARN source=ggml.go:152 msg="key not found" key=tokenizer.ggml.add_eot_token default=false
time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0
time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0
time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.num_channels default=0
time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.embedding_length default=0
time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.head_count default=0
time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0
time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0
time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.layer_norm_epsilon default=0
time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.local.freq_base default=10000
time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.global.freq_base default=1e+06
time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.freq_scale default=1
time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.mm_tokens_per_image default=256
time=2025-04-10T23:50:17.993+09:00 level=INFO source=server.go:405 msg="starting llama server" cmd="C:\Users\yukkuriTV\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --model C:\Users\yukkuriTV\.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2 --ctx-size 2048 --batch-size 512 --n-gpu-layers 15 --threads 6 --parallel 1 --port 50104"
time=2025-04-10T23:50:18.037+09:00 level=INFO source=sched.go:451 msg="loaded runners" count=1
time=2025-04-10T23:50:18.037+09:00 level=INFO source=server.go:580 msg="waiting for llama runner to start responding"
time=2025-04-10T23:50:18.042+09:00 level=INFO source=server.go:614 msg="waiting for server to become available" status="llm server error"
time=2025-04-10T23:50:18.068+09:00 level=INFO source=runner.go:816 msg="starting ollama engine"
time=2025-04-10T23:50:18.094+09:00 level=INFO source=runner.go:879 msg="Server listening on 127.0.0.1:50104"
time=2025-04-10T23:50:18.150+09:00 level=WARN source=ggml.go:152 msg="key not found" key=general.description default=""
time=2025-04-10T23:50:18.150+09:00 level=INFO source=ggml.go:67 msg="" architecture=gemma3 file_type=Q8_0 name=Gemma-3-4B-It description="" num_tensors=444 num_key_values=35
time=2025-04-10T23:50:18.295+09:00 level=INFO source=server.go:614 msg="waiting for server to become available" status="llm server loading model"
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: NVIDIA GeForce GTX 1650, compute capability 7.5, VMM: yes
load_backend: loaded CUDA backend from C:\Users\yukkuriTV\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
load_backend: loaded CPU backend from C:\Users\yukkuriTV\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
time=2025-04-10T23:50:19.230+09:00 level=INFO source=ggml.go:109 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.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-04-10T23:50:19.726+09:00 level=INFO source=ggml.go:289 msg="model weights" buffer=CPU size="3.1 GiB"
time=2025-04-10T23:50:19.726+09:00 level=INFO source=ggml.go:289 msg="model weights" buffer=CUDA0 size="1.4 GiB"
time=2025-04-10T23:50:22.780+09:00 level=INFO source=ggml.go:388 msg="compute graph" backend=CUDA0 buffer_type=CUDA0
time=2025-04-10T23:50:22.780+09:00 level=INFO source=ggml.go:388 msg="compute graph" backend=CPU buffer_type=CUDA_Host
time=2025-04-10T23:50:22.788+09:00 level=WARN source=ggml.go:152 msg="key not found" key=tokenizer.ggml.add_eot_token default=false
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.num_channels default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.embedding_length default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.head_count default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0
time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.layer_norm_epsilon default=0
time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.local.freq_base default=10000
time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.global.freq_base default=1e+06
time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.freq_scale default=1
time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.mm_tokens_per_image default=256
time=2025-04-10T23:50:22.809+09:00 level=INFO source=server.go:619 msg="llama runner started in 4.77 seconds"
[GIN] 2025/04/10 - 23:50:26 | 200 | 8.7700477s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:50:31 | 200 | 4.3203821s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:50:36 | 200 | 5.3622921s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:50:47 | 200 | 2.0825ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:50:56 | 200 | 1.6732ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:50:57 | 200 | 1.5642ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:51:09 | 200 | 1.5384ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:51:17 | 200 | 0s | 127.0.0.1 | GET "/api/version"
time=2025-04-10T23:51:34.541+09:00 level=INFO source=server.go:789 msg="llm predict error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input"
[GIN] 2025/04/10 - 23:51:34 | 200 | 40.6321ms | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:51:39 | 200 | 4.8736343s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:51:44 | 200 | 4.7278875s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:52:14 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:52:21 | 200 | 0s | 127.0.0.1 | GET "/api/version"
time=2025-04-10T23:52:59.251+09:00 level=INFO source=server.go:789 msg="llm predict error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input"
[GIN] 2025/04/10 - 23:52:59 | 200 | 51.8141ms | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:53:04 | 200 | 4.5159211s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:53:25 | 200 | 4.763401s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:55:50 | 200 | 2.5738ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:55:50 | 200 | 6.3µs | 127.0.0.1 | GET "/api/version"
time=2025-04-10T23:55:54.508+09:00 level=INFO source=server.go:789 msg="llm predict error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input"
[GIN] 2025/04/10 - 23:55:54 | 200 | 38.9413ms | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:55:59 | 200 | 4.3299013s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:56:03 | 200 | 4.4763269s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/04/10 - 23:56:57 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:58:36 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/10 - 23:58:43 | 200 | 3.0847ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:58:45 | 200 | 1.6163ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:58:59 | 200 | 1.5302ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:59:00 | 200 | 2.5908ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/10 - 23:59:05 | 200 | 0s | 127.0.0.1 | GET "/api/version"
time=2025-04-11T00:01:16.162+09:00 level=WARN source=sched.go:648 msg="gpu VRAM usage didn't recover within timeout" seconds=12.302142 model=C:\Users\yukkuriTV.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2
time=2025-04-11T00:01:16.375+09:00 level=WARN source=sched.go:648 msg="gpu VRAM usage didn't recover within timeout" seconds=12.5157752 model=C:\Users\yukkuriTV.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2
time=2025-04-11T00:01:16.630+09:00 level=WARN source=sched.go:648 msg="gpu VRAM usage didn't recover within timeout" seconds=12.7709461 model=C:\Users\yukkuriTV.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2
[GIN] 2025/04/11 - 00:02:14 | 200 | 2.7871ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/04/11 - 00:02:28 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2025/04/11 - 00:03:12 | 200 | 2.4986ms | 127.0.0.1 | GET "/api/tags"

<!-- gh-comment-id:2794198247 --> @yukkuriTV commented on GitHub (Apr 10, 2025): Sorry, here is the correct one. 2025/04/10 23:41:22 routes.go:1231: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION: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\\yukkuriTV\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-04-10T23:41:22.262+09:00 level=INFO source=images.go:458 msg="total blobs: 21" time=2025-04-10T23:41:22.264+09:00 level=INFO source=images.go:465 msg="total unused blobs removed: 0" time=2025-04-10T23:41:22.267+09:00 level=INFO source=routes.go:1298 msg="Listening on 127.0.0.1:11434 (version 0.6.5)" time=2025-04-10T23:41:22.269+09:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-04-10T23:41:22.269+09:00 level=INFO source=gpu_windows.go:167 msg=packages count=1 time=2025-04-10T23:41:22.269+09:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12 time=2025-04-10T23:41:22.535+09:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-23f2619b-2453-9ed8-f622-383f1a5428fb library=cuda compute=7.5 driver=12.6 name="NVIDIA GeForce GTX 1650" overhead="445.1 MiB" time=2025-04-10T23:41:22.538+09:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-23f2619b-2453-9ed8-f622-383f1a5428fb library=cuda variant=v12 compute=7.5 driver=12.6 name="NVIDIA GeForce GTX 1650" total="4.0 GiB" available="3.2 GiB" [GIN] 2025/04/10 - 23:43:48 | 200 | 77.7487ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:43:49 | 200 | 2.5884ms | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:47:31 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:48:16 | 200 | 4.151ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:48:16 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:48:31 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:48:37 | 200 | 7.7663ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:48:54 | 200 | 6.9112ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:48:55 | 200 | 2.0605ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:04 | 200 | 2.5982ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:05 | 200 | 12.318ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:13 | 200 | 1.5331ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:13 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:49:19 | 200 | 4.7153ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:29 | 200 | 1.5366ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:39 | 200 | 1.5306ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:40 | 200 | 1.531ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:47 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:49:53 | 200 | 1.5666ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:49:59 | 200 | 2.5731ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:50:00 | 200 | 2.0272ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:50:01 | 200 | 1.567ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:50:03 | 200 | 2.1286ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:50:07 | 200 | 0s | 127.0.0.1 | GET "/api/version" time=2025-04-10T23:50:17.896+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:17.896+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:17.897+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:17.897+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:17.908+09:00 level=INFO source=server.go:105 msg="system memory" total="11.8 GiB" free="5.3 GiB" free_swap="35.1 GiB" time=2025-04-10T23:50:17.908+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:17.908+09:00 level=INFO source=server.go:138 msg=offload library=cuda layers.requested=-1 layers.model=35 layers.offload=15 layers.split="" memory.available="[3.1 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.6 GiB" memory.required.partial="3.1 GiB" memory.required.kv="214.0 MiB" memory.required.allocations="[3.1 GiB]" memory.weights.total="3.8 GiB" memory.weights.repeating="3.2 GiB" memory.weights.nonrepeating="680.2 MiB" memory.graph.full="517.1 MiB" memory.graph.partial="1.0 GiB" time=2025-04-10T23:50:17.975+09:00 level=WARN source=ggml.go:152 msg="key not found" key=tokenizer.ggml.add_eot_token default=false time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0 time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0 time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.num_channels default=0 time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:17.978+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.embedding_length default=0 time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.head_count default=0 time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0 time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0 time=2025-04-10T23:50:17.979+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.layer_norm_epsilon default=0 time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.local.freq_base default=10000 time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.global.freq_base default=1e+06 time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.freq_scale default=1 time=2025-04-10T23:50:17.984+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.mm_tokens_per_image default=256 time=2025-04-10T23:50:17.993+09:00 level=INFO source=server.go:405 msg="starting llama server" cmd="C:\\Users\\yukkuriTV\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --model C:\\Users\\yukkuriTV\\.ollama\\models\\blobs\\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2 --ctx-size 2048 --batch-size 512 --n-gpu-layers 15 --threads 6 --parallel 1 --port 50104" time=2025-04-10T23:50:18.037+09:00 level=INFO source=sched.go:451 msg="loaded runners" count=1 time=2025-04-10T23:50:18.037+09:00 level=INFO source=server.go:580 msg="waiting for llama runner to start responding" time=2025-04-10T23:50:18.042+09:00 level=INFO source=server.go:614 msg="waiting for server to become available" status="llm server error" time=2025-04-10T23:50:18.068+09:00 level=INFO source=runner.go:816 msg="starting ollama engine" time=2025-04-10T23:50:18.094+09:00 level=INFO source=runner.go:879 msg="Server listening on 127.0.0.1:50104" time=2025-04-10T23:50:18.150+09:00 level=WARN source=ggml.go:152 msg="key not found" key=general.description default="" time=2025-04-10T23:50:18.150+09:00 level=INFO source=ggml.go:67 msg="" architecture=gemma3 file_type=Q8_0 name=Gemma-3-4B-It description="" num_tensors=444 num_key_values=35 time=2025-04-10T23:50:18.295+09:00 level=INFO source=server.go:614 msg="waiting for server to become available" status="llm server loading model" 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: NVIDIA GeForce GTX 1650, compute capability 7.5, VMM: yes load_backend: loaded CUDA backend from C:\Users\yukkuriTV\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll load_backend: loaded CPU backend from C:\Users\yukkuriTV\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll time=2025-04-10T23:50:19.230+09:00 level=INFO source=ggml.go:109 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.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-04-10T23:50:19.726+09:00 level=INFO source=ggml.go:289 msg="model weights" buffer=CPU size="3.1 GiB" time=2025-04-10T23:50:19.726+09:00 level=INFO source=ggml.go:289 msg="model weights" buffer=CUDA0 size="1.4 GiB" time=2025-04-10T23:50:22.780+09:00 level=INFO source=ggml.go:388 msg="compute graph" backend=CUDA0 buffer_type=CUDA0 time=2025-04-10T23:50:22.780+09:00 level=INFO source=ggml.go:388 msg="compute graph" backend=CPU buffer_type=CUDA_Host time=2025-04-10T23:50:22.788+09:00 level=WARN source=ggml.go:152 msg="key not found" key=tokenizer.ggml.add_eot_token default=false time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.num_channels default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.block_count default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.embedding_length default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.head_count default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.image_size default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.patch_size default=0 time=2025-04-10T23:50:22.791+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.vision.attention.layer_norm_epsilon default=0 time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.local.freq_base default=10000 time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.global.freq_base default=1e+06 time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.rope.freq_scale default=1 time=2025-04-10T23:50:22.795+09:00 level=WARN source=ggml.go:152 msg="key not found" key=gemma3.mm_tokens_per_image default=256 time=2025-04-10T23:50:22.809+09:00 level=INFO source=server.go:619 msg="llama runner started in 4.77 seconds" [GIN] 2025/04/10 - 23:50:26 | 200 | 8.7700477s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:50:31 | 200 | 4.3203821s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:50:36 | 200 | 5.3622921s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:50:47 | 200 | 2.0825ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:50:56 | 200 | 1.6732ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:50:57 | 200 | 1.5642ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:51:09 | 200 | 1.5384ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:51:17 | 200 | 0s | 127.0.0.1 | GET "/api/version" time=2025-04-10T23:51:34.541+09:00 level=INFO source=server.go:789 msg="llm predict error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input" [GIN] 2025/04/10 - 23:51:34 | 200 | 40.6321ms | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:51:39 | 200 | 4.8736343s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:51:44 | 200 | 4.7278875s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:52:14 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:52:21 | 200 | 0s | 127.0.0.1 | GET "/api/version" time=2025-04-10T23:52:59.251+09:00 level=INFO source=server.go:789 msg="llm predict error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input" [GIN] 2025/04/10 - 23:52:59 | 200 | 51.8141ms | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:53:04 | 200 | 4.5159211s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:53:25 | 200 | 4.763401s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:55:50 | 200 | 2.5738ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:55:50 | 200 | 6.3µs | 127.0.0.1 | GET "/api/version" time=2025-04-10T23:55:54.508+09:00 level=INFO source=server.go:789 msg="llm predict error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input" [GIN] 2025/04/10 - 23:55:54 | 200 | 38.9413ms | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:55:59 | 200 | 4.3299013s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:56:03 | 200 | 4.4763269s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/04/10 - 23:56:57 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:58:36 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/10 - 23:58:43 | 200 | 3.0847ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:58:45 | 200 | 1.6163ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:58:59 | 200 | 1.5302ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:59:00 | 200 | 2.5908ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/10 - 23:59:05 | 200 | 0s | 127.0.0.1 | GET "/api/version" time=2025-04-11T00:01:16.162+09:00 level=WARN source=sched.go:648 msg="gpu VRAM usage didn't recover within timeout" seconds=12.302142 model=C:\Users\yukkuriTV\.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2 time=2025-04-11T00:01:16.375+09:00 level=WARN source=sched.go:648 msg="gpu VRAM usage didn't recover within timeout" seconds=12.5157752 model=C:\Users\yukkuriTV\.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2 time=2025-04-11T00:01:16.630+09:00 level=WARN source=sched.go:648 msg="gpu VRAM usage didn't recover within timeout" seconds=12.7709461 model=C:\Users\yukkuriTV\.ollama\models\blobs\sha256-3d3a5470ffe8a3b9d5dff58f3be2c87c184b53f0b7fe25887328d17e8fd32eb2 [GIN] 2025/04/11 - 00:02:14 | 200 | 2.7871ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/04/11 - 00:02:28 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/04/11 - 00:03:12 | 200 | 2.4986ms | 127.0.0.1 | GET "/api/tags"
GiteaMirror commented 2026-04-29 02:23:46 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 10, 2025):

time=2025-04-10T23:50:17.908+09:00 level=INFO source=server.go:138 msg=offload library=cuda layers.requested=-1
 layers.model=35 layers.offload=15 layers.split="" memory.available="[3.1 GiB]" memory.gpu_overhead="0 B"
 memory.required.full="5.6 GiB" memory.required.partial="3.1 GiB" memory.required.kv="214.0 MiB"
 memory.required.allocations="[3.1 GiB]" memory.weights.total="3.8 GiB" memory.weights.repeating="3.2 GiB"
 memory.weights.nonrepeating="680.2 MiB" memory.graph.full="517.1 MiB" memory.graph.partial="1.0 GiB"

There is no mention of a projector here, so this model is not vision capable. Where did you get the model from?

<!-- gh-comment-id:2794231318 --> @rick-github commented on GitHub (Apr 10, 2025): ``` time=2025-04-10T23:50:17.908+09:00 level=INFO source=server.go:138 msg=offload library=cuda layers.requested=-1 layers.model=35 layers.offload=15 layers.split="" memory.available="[3.1 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.6 GiB" memory.required.partial="3.1 GiB" memory.required.kv="214.0 MiB" memory.required.allocations="[3.1 GiB]" memory.weights.total="3.8 GiB" memory.weights.repeating="3.2 GiB" memory.weights.nonrepeating="680.2 MiB" memory.graph.full="517.1 MiB" memory.graph.partial="1.0 GiB" ``` There is no mention of a projector here, so this model is not vision capable. Where did you get the model from?
GiteaMirror commented 2026-04-29 02:23:46 -05:00
Author
Owner
Copy Link

@yukkuriTV commented on GitHub (Apr 10, 2025):

https://huggingface.co/unsloth/gemma-3-4b-it-GGUF
This site.
https://huggingface.co/soob3123/amoral-gemma3-4B-v1/tree/main
Image recognition doesn't work on the model I got from here. However, image recognition isn't mentioned on the model's tag, so it may just not be included.

<!-- gh-comment-id:2794432221 --> @yukkuriTV commented on GitHub (Apr 10, 2025): https://huggingface.co/unsloth/gemma-3-4b-it-GGUF This site. https://huggingface.co/soob3123/amoral-gemma3-4B-v1/tree/main Image recognition doesn't work on the model I got from here. However, image recognition isn't mentioned on the model's tag, so it may just not be included.
GiteaMirror commented 2026-04-29 02:23:47 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 10, 2025):

https://huggingface.co/unsloth/gemma-3-4b-it-GGUF/discussions/4

<!-- gh-comment-id:2794435383 --> @rick-github commented on GitHub (Apr 10, 2025): https://huggingface.co/unsloth/gemma-3-4b-it-GGUF/discussions/4
GiteaMirror commented 2026-04-29 02:23:47 -05:00
Author
Owner
Copy Link

@megvadulthangya commented on GitHub (Apr 12, 2025):

I tried the same model, plus the 12B version... or hf.co/mradermacher/gemma3-4b-it-abliterated-GGUF:Q6_K or hf.co/mradermacher/Qwen2.5-VL-7B-Instruct-Vision-R1-GGUF:Q6_K
Same results... Visual function Does not work in ollama

I just say ...
hf.co/openbmb/MiniCPM-o-2_6-gguf:Q6_K its ok, but thats on ollama as well... :)

<!-- gh-comment-id:2798839022 --> @megvadulthangya commented on GitHub (Apr 12, 2025): I tried the same model, plus the 12B version... or hf.co/mradermacher/gemma3-4b-it-abliterated-GGUF:Q6_K or hf.co/mradermacher/Qwen2.5-VL-7B-Instruct-Vision-R1-GGUF:Q6_K Same results... Visual function Does not work in ollama _I just say ..._ hf.co/openbmb/MiniCPM-o-2_6-gguf:Q6_K _its ok, but thats on ollama as well... :)_
GiteaMirror commented 2026-04-29 02:23:47 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 12, 2025):

https://github.com/ollama/ollama/issues/10218#issuecomment-2794144223

<!-- gh-comment-id:2798839604 --> @rick-github commented on GitHub (Apr 12, 2025): https://github.com/ollama/ollama/issues/10218#issuecomment-2794144223
GiteaMirror commented 2026-04-29 02:23:47 -05:00
Author
Owner
Copy Link

@leporel commented on GitHub (Apr 14, 2025):

the same, want download Q6_K or Q5_K_M, but wherever I downloaded it from, it says vision is unsupported

ollama run hf.co/lmstudio-community/gemma-3-4b-it-GGUF:Q4_K_M
>>> G:\temp\test_1.png what you see?
Added image 'G:\temp\test_1.png'
Error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input

ollama run hf.co/bartowski/soob3123_amoral-gemma3-12B-GGUF:Q5_K_M
>>> G:\temp\test_1.png what you see?
I am unable to access local files, so I cannot tell you what is in the image file "G:\temp\test_1.png".
<!-- gh-comment-id:2802083004 --> @leporel commented on GitHub (Apr 14, 2025): the same, want download Q6_K or Q5_K_M, but wherever I downloaded it from, it says vision is unsupported ``` ollama run hf.co/lmstudio-community/gemma-3-4b-it-GGUF:Q4_K_M >>> G:\temp\test_1.png what you see? Added image 'G:\temp\test_1.png' Error: Failed to create new sequence: failed to process inputs: this model is missing data required for image input ``` ``` ollama run hf.co/bartowski/soob3123_amoral-gemma3-12B-GGUF:Q5_K_M >>> G:\temp\test_1.png what you see? I am unable to access local files, so I cannot tell you what is in the image file "G:\temp\test_1.png". ```
GiteaMirror commented 2026-04-29 02:23:48 -05:00
Author
Owner
Copy Link

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

If you want to use a q4_k_m quant with vision, https://ollama.com/library/gemma3:4b-it-q4_K_M

If you want a q5 or q6 quant with vision, download the fp16 model and create the quant you want:

ollama pull gemma3:12b-it-fp16
echo FROM gemma3:12b-it-fp16 > Modelfile
ollama create -q q6_k gemma3:12b-it-q6_k
<!-- gh-comment-id:2802110866 --> @rick-github commented on GitHub (Apr 14, 2025): If you want to use a q4_k_m quant with vision, https://ollama.com/library/gemma3:4b-it-q4_K_M If you want a q5 or q6 quant with vision, download the fp16 model and create the quant you want: ``` ollama pull gemma3:12b-it-fp16 echo FROM gemma3:12b-it-fp16 > Modelfile ollama create -q q6_k gemma3:12b-it-q6_k ```
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#53217
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?