Qwen3.6-35B-A3B

EXL3 3 bpw · sglang · 256K context · RTX 3080 · 10 GB

Qwen3.6's 35B mixture-of-experts model with 3B active parameters.

28 tok/sdecode

1,336 tok/sprefill

256Kcontext window

Tested on this cardOct 15, 2026

Run it

Follow the steps below to download and run this recipe. Omarchy Local AI lists the recipes it can launch on your hardware. The server then answers on http://localhost:8000/v1.

1 Download the weights

hf download yeasah/Qwen3.6-35B-A3B-exl3 \
  --revision 73da68ef827084d72c59936b4c2088c6580f4e2b \
  --local-dir ~/models/Qwen3.6-35B-A3B-exl3-73da68ef

2 Start the server

docker run --rm \
  --gpus all \
  -p 8000:30000 \
  --shm-size 16g \
  -e CUDA_DEVICE_ORDER=PCI_BUS_ID \
  -e EXL3_MOE_CPU_THREADS=8 \
  -e HF_HUB_OFFLINE=1 \
  -e SGLANG_EXL3_EMBED_HOST=1 \
  -e SGLANG_EXL3_EXPERT_CACHE_GB=auto \
  -e SGLANG_EXL3_MM_FAST_CPU=1 \
  -e SGLANG_EXL3_MODEL_PATH=/models/Qwen3.6-35B-A3B-exl3-73da68ef \
  -e SGLANG_EXL3_MOE_OFFLOAD=gpu_cache \
  -e SGLANG_EXL3_MOE_PREFILL_FP16_ACC=1 \
  -e SGLANG_EXL3_NGRAM_TIER=nvme \
  -e SGLANG_EXL3_OFFLOAD_COMPACT_PARTS=1 \
  -e SGLANG_EXL3_OFFLOAD_FUSED=1 \
  -e SGLANG_EXL3_OFFLOAD_STAGING_PARTS=4 \
  -e SGLANG_EXL3_VISION=0 \
  -e SGLANG_EXL3_VIT_MLP_CHUNK=8192 \
  -e SGLANG_EXL3_VIT_SDPA=1 \
  -v ~/models/Qwen3.6-35B-A3B-exl3-73da68ef:/models/Qwen3.6-35B-A3B-exl3-73da68ef:ro \
  --entrypoint /opt/entrypoint.sh \
  ghcr.io/0xsero/sglang-exl3-flashnext@sha256:ce47df25cee7ff235d8bfdf3e79e3828aa11bee3624db3ce121ebc79a8375dd3 \
  python3 \
  -m sglang.launch_server \
  --model-path /models/Qwen3.6-35B-A3B-exl3-73da68ef \
  --quantization exl3 \
  --trust-remote-code \
  --host 0.0.0.0 \
  --port 30000 \
  --served-model-name Qwen3.6-35B-A3B-exl3-73da68ef \
  --disable-shared-experts-fusion \
  --context-length 262144 \
  --mem-fraction-static 0.85 \
  --kv-cache-dtype fp8_e4m3 \
  --max-running-requests 1 \
  --chunked-prefill-size 4096 \
  --cuda-graph-max-bs-decode 4 \
  --cuda-graph-backend-prefill disabled \
  --max-mamba-cache-size 8 \
  --reasoning-parser qwen3 \
  --tool-call-parser qwen3_coder

What it passed

✓ loadLoads and serves within an hour

✓ chatAnswers a plain question and stops on its own

✓ reasoningThinks separately and gets 17 × 23 right

✓ toolsCalls a tool with the right arguments and uses the result

✓ contextRecalls a code buried in a prompt that fills 85% of the window

✓ speedDecodes at 15 tok/s or more

Passed all six checks on a real RTX 3080 on Oct 15, 2026.

Details

Weights: yeasah/Qwen3.6-35B-A3B-exl3 @ 73da68ef82 ›
Image: ghcr.io/0xsero/sglang-exl3-flashnext@sha256:ce47df25cee7ff235d8bfdf3e79e3828aa11bee3624db3ce121ebc79a8375dd3
Launch: sglang-qwen3.6-35b-a3b-exl3-offload-256k-rtx-3080-10gb ›
Recipe file: registry/recipes/nvidia/rtx-3080-10gb/qwen3.6-35b-a3b.sglang.256k.json ›
Model released: Apr 15, 2026
Tested: vast, RTX 3080