Model library / ornith-ai
Ornith-1.5-9B is a 9.7 billion-parameter dense model from ornith-ai that continues self-improvement through automated task generation and reinforcement learning.
Parameters in total
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
Ornith-1.5-9B builds on Ornith-1.0, extending its self-improvement loop to jointly optimise task generation, scaffold construction and solution rollouts. The publisher describes it as a foundation model that continuously creates new training tasks, discovers solving strategies and refines its policy without relying on a fixed human-curated task set.
The model contains 9.7 billion parameters and uses the Qwen3_5ForConditionalGeneration architecture with bfloat16 weights. It supports a context window of 262,144 tokens and is released under the MIT licence. Ornith-ai notes it is suited for single-GPU deployment and can be quantised for edge-device use.
Specification
| Repository | ornith-ai/Ornith-1.5-9B |
| Publisher | ornith-ai |
| Published | 18 August 2026 |
| Architecture | Qwen3_5ForConditionalGeneration |
| Model type | qwen3_5 |
| Parameters | 9,653,104,368 |
| Layers | 32 |
| Hidden size | 4,096 |
| Attention heads | 16, 4 key/value heads |
| Context length | 262,144 tokens |
| Vocabulary | 248,320 tokens |
| Weight format | bfloat16 |
| Licence | mit |
Memory
Parameter count multiplied by the bytes each format uses. The key-value cache comes on top.
| Precision | Weights | Notes |
|---|---|---|
| bf16 | 18.0 GiB | Full precision as released |
| fp8 | 9.0 GiB | 8-bit, near-lossless on most models |
| int4 | 4.5 GiB | 4-bit, smallest footprint |
The cache costs 128 KiB per token at 16-bit, so the full 262,144-token context of one request needs about 32.0 GiB. Concurrency multiplies that number, not the weights.
Hardware
At the released precision, with 90% of the card given to the server and 2.5 GiB kept for the runtime. The context column is what is left for the cache on that setup.
| GPU | Cards needed | Free for cache | Context it holds |
|---|---|---|---|
| RTX 4090 24GB | 1 | 1.1 GiB | about 9,172 tokens |
| RTX 5090 32GB | 1 | 8.3 GiB | about 68,154 tokens |
| L40S 48GB | 1 | 22.7 GiB | about 186,119 tokens |
| A100 80GB | 1 | 51.5 GiB | the full 262,144 tokens |
| H100 80GB | 1 | 51.5 GiB | the full 262,144 tokens |
| RTX PRO 6000 96GB | 1 | 65.9 GiB | the full 262,144 tokens |
| H200 141GB | 1 | 106.4 GiB | the full 262,144 tokens |
Serving
A vLLM launch line for the shape above. Check the model card for a runtime the publisher recommends.
vllm serve ornith-ai/Ornith-1.5-9B \ --max-model-len 8192 \ --gpu-memory-utilization 0.90
Questions
The weights take about 18 GiB at the released precision. That fits on one RTX 4090 24GB, which leaves roughly 1 GiB for the key-value cache.
The repository declares mit. Read the licence text before commercial use: the name alone does not tell you what is allowed.
The configuration allows 262,144 tokens. Whether the whole window is usable depends on the memory left for the cache, which the hardware table works out per card.
This page is a reference, and Aiqre does not serve this model today. We deploy open-weight models on dedicated EU hardware on request, so if you want an OpenAI-compatible endpoint for it, email [email protected] with the model name and roughly what volume you expect.
Also in the library
Ornith-1.5-35B-A3B is a 36-billion-parameter mixture-of-experts language model released by ornith-ai.
ZGCM-1 is a 7.4 billion-parameter dense language model designed for mathematical reasoning and tool-assisted search.
NeoHorse-1-9B is a 9 billion-parameter causal language model fine-tuned for text-based agent tasks, tool use, coding and instruct…
LLaDA2.2-mini is a 16.3 billion-parameter MoE diffusion language model with 128 k token context, supporting Levenshtein editing.
Hosting
Tell us the model and roughly what volume you expect. We reply with a price and an OpenAI-compatible endpoint, with your prompts never stored.