no model found with name while using the model

[Hello-World-Traveler]

LocalAI version:
3.1.1
Environment, CPU architecture, OS, and Version:
docker

Describe the bug
ERR failed to install model "bark-small_weights-f16.bin" from gallery error="no model found with name \"bark-small_weights-f16.bin\"" when already installed.

I was chatting away with deepseek then nothing, check log:
ERR failed to install model "deepseek-r1-distill-llama-8b-Q4_K_M.gguf" from gallery error="no model found with name \"deepseek-r1-distill-llama-8b-Q4_K_M.gguf\"" once again it's already installed.

To Reproduce
Not sure

Expected behavior
answers question

Logs

9:36AM DBG Stream request received

9:36AM INF Success ip= latency=393.902938ms method=POST status=200 url=/v1/chat/completions

9:36AM DBG Sending chunk: {"created":1751189799,"object":"chat.completion.chunk","id":"3aff22668","model":"deepseek-r1-distill-llama-8b","choices":[{"index":0,"finish_reason":"","delta":{"role":"assistant","content":""}}],"usage":{"prompt_tokens":0,"completion_tokens":0,"total_tokens":0}}

9:36AM ERR failed to install model "deepseek-r1-distill-llama-8b-Q4_K_M.gguf" from gallery error="no model found with name \"deepseek-r1-distill-llama-8b-Q4_K_M.gguf\""

9:43AM ERR failed to install model "bark-small_weights-f16.bin" from gallery error="no model found with name \"bark-small_weights-f16.bin\""

9:43AM DBG Stopping all backends except 'bark-cpp-small'

9:43AM DBG Deleting process deepseek-r1-distill-llama-8b

9:43AM INF BackendLoader starting backend=bark-cpp modelID=bark-cpp-small o.model=bark-small_weights-f16.bin

9:43AM DBG Loading model in memory from file: /models/bark-small_weights-f16.bin

9:43AM DBG Loading Model bark-cpp-small with gRPC (file: /models/bark-small_weights-f16.bin) (backend: bark-cpp): {backendString:bark-cpp model:bark-small_weights-f16.bin modelID:bark-cpp-small assetDir:/tmp/localai/backend_data context:{emptyCtx:{}} gRPCOptions:0xc0be85e588 externalBackends:map[bark:/build/backend/python/bark/run.sh coqui:/build/backend/python/coqui/run.sh diffusers:/build/backend/python/diffusers/run.sh exllama2:/build/backend/python/exllama2/run.sh faster-whisper:/build/backend/python/faster-whisper/run.sh kokoro:/build/backend/python/kokoro/run.sh rerankers:/build/backend/python/rerankers/run.sh transformers:/build/backend/python/transformers/run.sh vllm:/build/backend/python/vllm/run.sh] grpcAttempts:20 grpcAttemptsDelay:2 parallelRequests:false}

9:43AM DBG Loading external backend: /backends/bark-cpp/run.sh

9:43AM DBG external backend is file: &{name:run.sh size:191 mode:448 modTime:{wall:389370631 ext:63885975675 loc:0x468dc20} sys:{Dev:66306 Ino:5126488 Nlink:1 Mode:33216 Uid:0 Gid:0 X__pad0:0 Rdev:0 Size:191 Blksize:4096 Blocks:8 Atim:{Sec:1751189621 Nsec:120067606} Mtim:{Sec:1750378875 Nsec:389370631} Ctim:{Sec:1751189621 Nsec:118067467} X__unused:[0 0 0]}}

9:43AM DBG Loading GRPC Process: /backends/bark-cpp/run.sh

9:43AM DBG GRPC Service for bark-cpp-small will be running at: '127.0.0.1:35967'

9:43AM DBG GRPC Service state dir: /tmp/go-processmanager2135792008

9:43AM DBG GRPC Service Started

9:43AM DBG Wait for the service to start up

9:43AM DBG Options: ContextSize:1024 Seed:1117498056 NBatch:512 MMap:true NGPULayers:9999999 Threads:10

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stdout Initializing libbackend for bark-cpp

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stdout virtualenv activated

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stdout activated virtualenv has been ensured

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stderr Traceback (most recent call last):

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stderr   File "/backends/bark-cpp/backend.py", line 11, in <module>

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stderr     from scipy.io.wavfile import write as write_wav

9:43AM DBG GRPC(bark-cpp-small-127.0.0.1:35967): stderr ModuleNotFoundError: No module named 'scipy'

Additional context
It looks like Local-ai is trying to open another backend, and closing the one that is in use.
Restart fixes the issue but this could be a temp fix. Might have to use the active backend option.

[alixanderthegreat]

I think I found a component of this: it is using the option of //models

sudo docker run -ti --name local-ai -p 8080:8080 -v $PWD/models:/models localai/localai:latest
[...opening sequence...]
1:51PM INF Setting logging to info
1:51PM INF Starting LocalAI using 4 threads, with models path: //models
1:51PM INF LocalAI version: v3.1.1 (cd2b0c0e7cfbbb55f1b4b6b2d4c76d2962bc8fc5)
1:51PM INF Preloading models from //models
1:51PM INF core/startup process completed!

And then after installing a model

2:05PM INF Success ip=172.17.0.1 latency=14.407412ms method=POST status=200 url=/browse/install/model/localai@gemma-3-1b-it
2:05PM INF Downloading "https://huggingface.co/ggml-org/gemma-3-1b-it-GGUF/resolve/main/gemma-3-1b-it-Q4_K_M.gguf"
2:05PM INF Success ip=172.17.0.1 latency="480.758µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="565.967µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="52.147µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="620.92µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="44.653µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="47.078µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="45.635µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Downloading /models/gemma-3-1b-it-Q4_K_M.gguf.partial: 203.2 MiB/768.7 MiB (26.44%) ETA: 13.921576488s
2:05PM INF Success ip=172.17.0.1 latency="141.013µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="45.825µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency=3.040139ms method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="43.852µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="65.553µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="46.397µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="40.385µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="40.375µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="41.908µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Downloading /models/gemma-3-1b-it-Q4_K_M.gguf.partial: 456.7 MiB/768.7 MiB (59.41%) ETA: 6.916499845s
2:05PM INF Success ip=172.17.0.1 latency="51.396µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="44.663µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="51.446µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="56.385µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="55.244µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="38.943µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="44.723µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="50.304µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Downloading /models/gemma-3-1b-it-Q4_K_M.gguf.partial: 677.0 MiB/768.7 MiB (88.07%) ETA: 2.049638485s
2:05PM INF Success ip=172.17.0.1 latency="46.957µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="56.105µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency="43.521µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF File "/models/gemma-3-1b-it-Q4_K_M.gguf" downloaded and verified
2:05PM INF Preloading models from //models
                                    

  Model name: gemma-3-1b-it                                                   


2:05PM INF Success ip=172.17.0.1 latency="813.329µs" method=GET status=200 url=/browse/job/progress/221ea520-54f2-11f0-9786-b2dd09551b5f
2:05PM INF Success ip=172.17.0.1 latency=5.440773ms method=GET status=200 url=/browse/job/221ea520-54f2-11f0-9786-b2dd09551b5f
2:06PM INF Success ip=172.17.0.1 latency=41.476051ms method=GET status=200 url=/chat/

[...]

2:06PM INF Success ip=172.17.0.1 latency=5.786704508s method=POST status=200 url=/v1/chat/completions
2:06PM ERR failed to install model "gemma-3-1b-it-Q4_K_M.gguf" from gallery error="no model found with name \"gemma-3-1b-it-Q4_K_M.gguf\""
2:06PM INF Trying to load the model 'gemma-3-1b-it' with the backend '[llama-cpp llama-cpp-fallback stablediffusion-ggml whisper piper silero-vad huggingface]'
2:06PM INF [llama-cpp] Attempting to load
2:06PM INF BackendLoader starting backend=llama-cpp modelID=gemma-3-1b-it o.model=gemma-3-1b-it-Q4_K_M.gguf
2:06PM INF [llama-cpp] attempting to load with AVX2 variant
2:06PM INF [llama-cpp] Loads OK

However, the bot does respond in the ui:

You
hey there bot

gemma-3-1b-it
Hey there! How's your day going so far? 😊

What can I do for you today? Do you have any questions, or would you like to chat about something?

I went and looked into the Dockerfile and it looks like line 360 adds a WORKDIR at /

[Hello-World-Traveler]

It used to be /build I think, there is a fix for next update, I hope..
it's a pain to restart every few minutes after use.

[Hello-World-Traveler]

I went back to LocalAI Version 18c3dc3 (before 3.0), as it kept trying to open an unneeded backend and close the one I was using. After a few restarts I was still unable to use it.