#412 V0.1.32 Vision Encoder load error (Okay in v0.1.31)

Beschreibung

.32 strata-coder-iq1_m.log is 0 bytes

Win 11, 64GB RAM amd 5950x, RTX5090

Let me know if you need any more info.


Error at:-------------------------------------------------------------------------------------

loading the vision encoder ...
Traceback (most recent call last):
  File "F:\Utils\AI-01\Strata-0.1.32\serve\server.py", line 2546, in <module>
    sys.exit(main())
             ~~~~^^
  File "F:\Utils\AI-01\Strata-0.1.32\serve\server.py", line 2415, in main
    vision = Vision(vcfg, log=open(cfg["log"], "a", encoding="utf-8") if cfg.get("log") else None, env=env)
  File "F:\Utils\AI-01\Strata-0.1.32\serve\server.py", line 508, in __init__
    self._start()
    ~~~~~~~~~~~^^
  File "F:\Utils\AI-01\Strata-0.1.32\serve\server.py", line 519, in _start
    raise RuntimeError("the vision encoder did not start: " + line.strip())
RuntimeError: the vision encoder did not start:


Preamble: ------------------------------------------------------------------------------

Strata - Qwen3.8-Flash-Next on a normal PC (a GPU + system RAM + CPU)
  Found your earlier install in F:\Utils\AI-01\Strata-0.1.31 (coder-iq1_m): setting up this copy the same way - the model files are reused, nothing big is downloaded.

=== Step 1: checking your PC ===
  [ok] GPU: NVIDIA GeForce RTX 5090, 31.8 GB VRAM, compute capability 12.0, driver 616.92
  [ok] RAM: 64 GB
  [ok] CPU: AMD Ryzen 9 5950X 16-Core Processor (AVX2)

=== Step 2: your choices ===
  [ok] model: Qwen3.8-Flash-Next Coder

  1) IQ1_M    the Coder's only size: half the experts, stored like IQ3_S (3.5 bits); download 58 GB, uses ~23 GB of RAM
  [ok] size: IQ1_M
  [ok] context: 131072 tokens
  [ok] KV cache: 8-bit
  [ok] images: on
  [ok] experimental speed projection: off

=== Step 3: Python packages ===
  Installing numpy, jinja2, regex, pyyaml, tqdm, requests, cmake, ninja, pillow, psutil ...
  > F:\Utils\AI-01\Strata-0.1.32\.venv\Scripts\python.exe -m pip install --quiet --disable-pip-version-check numpy==2.5.3; python_version >= "3.12" numpy==2.4.6; python_version == "3.11" numpy==2.2.6; python_version < "3.11" jinja2==3.1.6 regex==2026.9.10 pyyaml==6.0.3 tqdm==4.70.1 requests==2.34.2 cmake==4.4.3 ninja==1.13.2 pillow==12.3.0 psutil==7.2.2 markupsafe==3.0.3 certifi==2026.7.22 charset-normalizer==3.5.1 idna==3.20 urllib3==2.8.0 colorama==0.4.6; sys_platform == "win32"
  [ok] numpy, jinja2, regex, pyyaml, tqdm, requests, cmake, ninja, pillow, psutil installed

=== Step 4: the Strata engine ===
  llama.cpp source:    33.6 MB
  [ok] llama.cpp source downloaded
  [ok] llama.cpp 3cf0325 (gguf-py, ggml, mtmd)
  Downloading the ready-made Strata engine ...
  Strata engine:   0.11 / 0.12 GB (88%)
  [ok] Strata engine downloaded
  [ok] ready-made engine 0.1.32 for sm_75, sm_86, sm_89, sm_120 (CUDA 13.0)
  Installing NVIDIA CUDA libraries (cuBLAS, CUDA runtime; ~0.4 GB) ...
  > F:\Utils\AI-01\Strata-0.1.32\.venv\Scripts\python.exe -m pip install --quiet --disable-pip-version-check nvidia-cublas==13.0.2.14 nvidia-cuda-runtime==13.0.96
  [ok] NVIDIA CUDA libraries (cuBLAS, CUDA runtime; ~0.4 GB) installed
  [ok] engine: F:\Utils\AI-01\Strata-0.1.32\engine\strata.exe

=== Step 5: downloading Qwen3.8-Flash-Next Coder IQ1_M ===
  [ok] Qwen3.8-Flash-Next-GSQ-RCO-IQ1_M-00001-of-00002.gguf already downloaded
  [ok] Qwen3.8-Flash-Next-GSQ-RCO-IQ1_M-00002-of-00002.gguf already downloaded
  [ok] model files present
  [ok] vision encoder already downloaded
  [ok] vision encoder: F:\Utils\AI-01\Strata-data\models\mmproj-Qwen3.8-Flash-Next-BF16.gguf

=== Step 6: preparing the model for Strata ===
  [ok] model prepared: F:\Utils\AI-01\Strata-data\packs\coder-iq1_m
  [ok] MTP draft layer: F:\Utils\AI-01\Strata-data\mtp\rt

=== Step 7: writing the start script ===
  [ok] KV streaming on: the context's KV cache lives in RAM (1.8 GB), more experts fit in VRAM
  [ok] the settings tuned for this PC earlier are used (2026-09-29)
  [ok] start script: run-coder-iq1_m.bat

All set.
  API (OpenAI):     http://127.0.0.1:8080/v1   (any API key; model name: anything)
  API (Anthropic):  http://127.0.0.1:8080/v1/messages
  Next time:        just run START-HERE.bat (or run-coder-iq1_m.bat) - it starts right away
  Images:           send them in the chat page, in chat.py (/image <path>) or over the API
  [ok] GPU: GPU 0 (NVIDIA GeForce RTX 5090, 32 GB)

  ----------------------------------------------------------------------------------------------------
  Starting qwen3.8-flash-next-coder-iq1_m: it loads about 30 GB into RAM and locks part of it for the GPU.
  While it does, YOUR PC CAN BE SLOW OR STOP RESPONDING FOR 1-3 MINUTES (longer the first time after a
  restart). That is normal: please wait and don't close this window - the browser opens when it is ready.
  Later, closing this window stops the model.
  ----------------------------------------------------------------------------------------------------
loading the vision encoder ...