Ornith-1.5-35B-A3B

NVFP4 · sglang · 128K context · RTX PRO 4500 Blackwell · 32 GB

Ornith 1.5, a 35B mixture-of-experts with built-in MTP.

– tok/sdecode

– tok/sprefill

128Kcontext window

Earlier checkearlier

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 ornith-ai/Ornith-1.5-35B-A3B-NVFP4 \
  --revision 0f0b1b59b879ccde1353e6ebd0fb10c204d4c544

2 Start the server

docker run --rm \
  --gpus all \
  -p 8000:30000 \
  --shm-size 16g \
  -e HF_HOME=/root/.cache/huggingface \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  --entrypoint /opt/sglang/bin/python \
  lmsysorg/sglang:dev-cu13@sha256:6cd4635214f279e0a43019f88e3120d407567640a58aa7dcc0085e3d91402cc4 \
  -m sglang.launch_server \
  --model-path ornith-ai/Ornith-1.5-35B-A3B-NVFP4 \
  --revision 0f0b1b59b879ccde1353e6ebd0fb10c204d4c544 \
  --tp 1 \
  --host 0.0.0.0 \
  --port 30000 \
  --context-length 131072 \
  --mem-fraction-static 0.90 \
  --max-running-requests 4 \
  --cuda-graph-max-bs-decode 4 \
  --enable-cache-report \
  --trust-remote-code \
  --reasoning-parser qwen3 \
  --tool-call-parser qwen3_coder \
  --kv-cache-dtype fp8_e4m3 \
  --attention-backend flashinfer \
  --moe-runner-backend flashinfer_cutlass

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 the older check (loads and chats); a full six-check run is pending.

Details

Weights: ornith-ai/Ornith-1.5-35B-A3B-NVFP4 @ 0f0b1b59b8 ›
Image: lmsysorg/sglang:dev-cu13@sha256:6cd4635214f279e0a43019f88e3120d407567640a58aa7dcc0085e3d91402cc4
Launch: sglang-ornith-1.5-35b-a3b-nvfp4-128k-v2 ›
Recipe file: registry/recipes/nvidia/rtx-pro-4500-blackwell-32gb/ornith-1.5-35b-a3b.sglang.128k.json ›
Model released: Aug 18, 2026
Tested: earlier acceptance