Nanbeige4.3-33B
NVFP4 · vllm · 256K context · RTX PRO 6000 Blackwell · 96 GB
Nanbeige4.3, a 33B dense Llama-architecture model with thinking, tool calls and a 512K window.
71 tok/sdecode
1,505 tok/sprefill
256Kcontext window
Tested on this cardOct 30, 2026
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 primitive-ai/Nanbeige4.3-33B-NVFP4 \ --revision 5ddf8788ed1fa4bec0c0b6fc74ddb0bef7dc4212 \ --local-dir ~/models/Nanbeige4.3-33B-NVFP4-5ddf8788
- 2 Start the server
docker run --rm \ --gpus all \ -p 8000:8000 \ --shm-size 16g \ -v ~/models/Nanbeige4.3-33B-NVFP4-5ddf8788:/models/Nanbeige4.3-33B-NVFP4-5ddf8788:ro \ --entrypoint vllm \ vllm/vllm-openai:v0.30.0@sha256:8a69ffad015f138d7170c4ddc429e230a3bc1c1719f67e14324749df200a4b90 \ serve \ /models/Nanbeige4.3-33B-NVFP4-5ddf8788 \ --served-model-name Nanbeige4.3-33B-NVFP4-5ddf8788 \ --host 0.0.0.0 \ --port 8000 \ --max-model-len 262144 \ --max-num-seqs 1 \ --gpu-memory-utilization 0.9 \ --kv-cache-dtype fp8 \ --enable-auto-tool-choice \ --tool-call-parser qwen3_coder \ --reasoning-parser qwen3
✓ 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 PRO 6000 Blackwell WS (450 W cap) on Oct 30, 2026.
- Weights
- primitive-ai/Nanbeige4.3-33B-NVFP4 @ 5ddf8788ed ›
- Image
- vllm/vllm-openai:v0.30.0@sha256:8a69ffad015f138d7170c4ddc429e230a3bc1c1719f67e14324749df200a4b90
- Engine profile
- vllm ›
- Recipe file
- registry/recipes/nvidia/rtx-pro-6000-blackwell-96gb/nanbeige4.3-33b.vllm.256k.json ›
- Model released
- Oct 24, 2026
- Tested
- owner, RTX PRO 6000 Blackwell WS (450 W cap)