Model library / ornith-ai
Ornith-1.5-122B is a 125.1 billion-parameter mixture-of-experts language model from ornith-ai, released under the MIT licence.
Mixture of experts, count not published
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
Ornith-1.5-122B is a 125.1 billion-parameter mixture-of-experts model built on Qwen-3.5 and Gemma-4, extending Ornith-1.0 with a joint optimisation loop for task generation, scaffold construction and solution rollouts. According to ornith-ai it is stronger, faster and leaner, offering improved instruction following, task decomposition and real-world knowledge while running on a single DGX.
In practice the model occupies 125.1 billion parameters in a mixture-of-experts architecture, uses bfloat16 weights and supports a 262 144-token context window. It is released under the MIT licence, can be deployed on a single GPU such as a DGX, and has a quantised mobile variant for edge devices.
Specification
| Repository | ornith-ai/Ornith-1.5-122B |
| Publisher | ornith-ai |
| Published | 16 October 2026 |
| Architecture | Qwen3_5MoeForConditionalGeneration |
| Model type | qwen3_5_moe |
| Parameters | 125,086,497,008 |
| Layers | 48 |
| Hidden size | 3,072 |
| Attention heads | 32, 2 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 | 233.0 GiB | Full precision as released |
| fp8 | 116.5 GiB | 8-bit, near-lossless on most models |
| int4 | 58.2 GiB | 4-bit, smallest footprint |
The cache costs 96 KiB per token at 16-bit, so the full 262,144-token context of one request needs about 24.0 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 | more than 8 | — | — |
| RTX 5090 32GB | more than 8 | — | — |
| L40S 48GB | 8 | 92.6 GiB | the full 262,144 tokens |
| A100 80GB | 4 | 45.0 GiB | the full 262,144 tokens |
| H100 80GB | 4 | 45.0 GiB | the full 262,144 tokens |
| RTX PRO 6000 96GB | 4 | 102.6 GiB | the full 262,144 tokens |
| H200 141GB | 2 | 15.8 GiB | about 172,668 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-1.5-122B \ --tensor-parallel-size 8 \ --max-model-len 262144 \ --gpu-memory-utilization 0.90
Questions
The weights take about 233 GiB at the released precision. No single card in the table above holds that, so it needs several GPUs or a lower precision.
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.
Ornith-1.5-9B is a 9.7 billion-parameter dense model from ornith-ai that continues self-improvement through automated task genera…
Kolibri is a 78.1 billion-parameter mixture-of-experts language model from Aleph-Alpha, supporting German and English with long-c…
Computer-10 is a 70-billion-parameter Llama 3.1-based document model that generates conversational turns without instruction tuni…
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.