Ornith-1.5-9B

EXL3 4 bpw · tabbyapi · 128K context · RTX 3090 · 24 GB

Ornith 1.5, the dense 9B.

– tok/sdecode

– tok/sprefill

128Kcontext window

Earlier checkSep 22, 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 ultimatechris/Ornith-1.5-9B-EXL3-4bpw \
  --revision 8d1e89ef430f9ba3a12789b198abf1385670ceb0

2 Write the server config

cat > ornith-1.5-9b.tabbyapi.128k.yml <<'EOF'
network:
  host: 0.0.0.0
  port: 5000
  disable_auth: true
  disable_fetch_requests: true
  send_tracebacks: false
  api_servers: ["OAI"]
  sse_ping_interval: 15

logging:
  log_prompt: false
  log_generation_params: false
  log_requests: false
  log_chat_completion_requests: false

model:
  model_dir: /opt/models
  inline_model_loading: false
  use_dummy_models: false
  model_name: Ornith-1.5-9B-EXL3-4bpw
  backend: exllamav3
  max_seq_len: 131072
  cache_size: 132096
  cache_mode: Q4
  tensor_parallel: false
  gpu_split_auto: true
  autosplit_reserve: [512]
  chunk_size: 1024
  output_chunking: true
  max_batch_size: 2
  vision: true
  vision_offload: true
  reasoning: true
  start_in_reasoning: always
  reasoning_start_token: " thinking"
  reasoning_end_token: "</think>"
  template_vars_default:
    enable_thinking: true
  tool_format: qwen3_coder

draft_model:
  draft_mode: disabled

sampling:
  override_preset:

memory:
  sysmem_recurrent_cache: 4096
  sysmem_kv_cache: 0
  cuda_malloc_async: true
EOF

3 Start the server

docker run --rm \
  --gpus all \
  -p 8000:5000 \
  --ipc host \
  --shm-size 16g \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  -v $PWD/ornith-1.5-9b.tabbyapi.128k.yml:/app/config.yml:ro \
  --entrypoint /opt/venv/bin/python3 \
  ghcr.io/theroyallab/tabbyapi@sha256:1c13dd17416660856e9962aeee8498827f0acd604eb36795128a03ddfe03c6ef \
  main.py \
  --config /app/config.yml

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: ultimatechris/Ornith-1.5-9B-EXL3-4bpw @ 8d1e89ef43 ›
Image: ghcr.io/theroyallab/tabbyapi@sha256:1c13dd17416660856e9962aeee8498827f0acd604eb36795128a03ddfe03c6ef
Launch: tabbyapi-ornith-1.5-9b-4bpw-128k ›
Recipe file: registry/recipes/nvidia/rtx-3090-24gb/ornith-1.5-9b.tabbyapi.128k.json ›
Model released: Aug 18, 2026
Tested: earlier acceptance