Leaderboard / Ornith

Ornith 1.5 9B

Ornith · Ornith 1.5 · released 2026-08-18 · data checked 2026-11-13

Ornith 1.5 9B is a 9.7B-parameter dense model from Ornith, released 2026-08-18 under the MIT license. It scores 72.9 on the Small LM Index, #1 of 31 small models. It needs about 6.8 GB of memory at 4-bit with an 8K-token context.

Further trained from Qwen3.5 9B with a self-improvement loop that generates its own training tasks and agent scaffolds. Tuned for terminal coding agents. Scores are the mean of five runs.

Small LM Index
72.9
#1 of 31 scored small models
Likely 66.8–77.8 · from 3 of 6 core benchmarks
Parameters
9.7B
Dense
Price per 1M tokens
—
No hosted API on OpenRouter; self-host
Context window
262K
tokens

Benchmarks

Each bar shows this model's score. The tick marks the best small model on that benchmark. Sources: 4 reported by Ornith. How we label sources.

GPQA Diamond Lab-reported Knowledge86.4
Humanity's Last Exam Lab-reported Reasoning · No tools (with tools: 30.5)20.2
SWE-bench Verified Lab-reported Coding · OpenHands harness70.6
Terminal-Bench Lab-reported Agentic · Terminal-Bench 2.1 (Terminus 2)46.2

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

MMLU-Pro≈ 86
AIME≈ 97
LiveCodeBench≈ 89

Specifications

Lab
Ornith (country not published · Other)
Based on
Qwen3.5 9B (fine-tune or distillation)
Architecture
Dense transformer
Total parameters
9.7B
Active parameters
All (dense)
Size class
Small
Context window
262K tokens
Input modalities
text, image
Reasoning
Reasoning (thinking)
License
MIT
Openness
Open weights
Memory to run
~6.8 GB at 4-bit, 8K context (more sizes)
Hugging Face
582.5K downloads (30 days) · 421 likes

Compare with

Ranked in

Sources

Something wrong here? Report it.

Run it locally

Uses bartowski/Ornith-1.5-9B-GGUF (Q4_K_M). Swap the tag for another size: BF16, IQ2_M, IQ3_M, IQ3_XS, IQ3_XXS, IQ4_NL, IQ4_XS, Q2_K….

Ollama
ollama run hf.co/bartowski/Ornith-1.5-9B-GGUF:Q4_K_M
llama.cpp
llama-server -hf bartowski/Ornith-1.5-9B-GGUF:Q4_K_M
Then open http://localhost:8080 for a chat window.
vLLM (GPU server)
vllm serve ornith-ai/Ornith-1.5-9B
Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization4K context32K context256K context
16-bit F1620 GB21 GB29 GB
8-bit Q8_011 GB12 GB20 GB
4-bit Q4_K_M6.7 GB7.6 GB15 GB

Check it against your GPU, Mac or phone →