Model library / ai9stars
G9v3-39A5B is a 39-billion-parameter Mixture-of-Experts causal language model for assistant, coding, tool-use and reasoning tasks.
32 of 320 experts active per token
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
G9v3-39A5B is a 39-billion-parameter Mixture-of-Experts causal language model from AI9Stars, designed for everyday assistant tasks, coding, tool-use workflows and reasoning, supporting both Think and No Think modes from a single checkpoint.
The model uses 320 experts with 32 active per token, giving roughly 5 billion activated parameters per token. It accepts up to 131,072 tokens of context, runs in bfloat16 weight format and is released under the Apache-2.0 licence, making it suitable for local or self-hosted deployment.
Specification
| Repository | ai9stars/G9v3-39A5B |
| Publisher | ai9stars |
| Published | 21 July 2026 |
| Architecture | G9v3ForCausalLM |
| Model type | g9v3 |
| Parameters | 38,967,481,920 |
| Experts | 320 total, 32 active per token |
| Layers | 38 |
| Hidden size | 2,048 |
| Attention heads | 32, 2 key/value heads |
| Context length | 131,072 tokens |
| Vocabulary | 130,560 tokens |
| Weight format | bfloat16 |
| Licence | apache-2.0 |
Memory
Parameter count multiplied by the bytes each format uses. The key-value cache comes on top.
| Precision | Weights | Notes |
|---|---|---|
| bf16 | 72.6 GiB | Full precision as released |
| fp8 | 36.3 GiB | 8-bit, near-lossless on most models |
| int4 | 18.1 GiB | 4-bit, smallest footprint |
The cache costs 38 KiB per token at 16-bit, so the full 131,072-token context of one request needs about 4.8 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 | 4 | 3.8 GiB | about 105,337 tokens |
| RTX 5090 32GB | 4 | 32.6 GiB | the full 131,072 tokens |
| L40S 48GB | 2 | 8.8 GiB | the full 131,072 tokens |
| A100 80GB | 2 | 66.4 GiB | the full 131,072 tokens |
| H100 80GB | 2 | 66.4 GiB | the full 131,072 tokens |
| RTX PRO 6000 96GB | 1 | 11.3 GiB | the full 131,072 tokens |
| H200 141GB | 1 | 51.8 GiB | the full 131,072 tokens |
Serving
A vLLM launch line for the shape above. Check the model card for a runtime the publisher recommends.
vllm serve ai9stars/G9v3-39A5B \ --max-model-len 102400 \ --gpu-memory-utilization 0.90
Questions
The weights take about 73 GiB at the released precision. That fits on one RTX PRO 6000 96GB, which leaves roughly 11 GiB for the key-value cache.
The repository declares apache-2.0. Read the licence text before commercial use: the name alone does not tell you what is allowed.
The configuration allows 131,072 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 hello@aiqre.com with the model name and roughly what volume you expect.
Also in the library
Hy-Memory-30B-A3B is a Tencent-developed long-term-memory writer model designed to replace LLMs in Mem0, MemOS and Hy-Memory fram…
Granite Switch 4.1 30B is a modular LLM combining a 30-billion base, decision and guardian adapters, and an English speech-to-tex…
Sabi-Noema-MLA is a 27.4 B-parameter agentic model derived from Qwen3.8-27B, using multi-head latent attention to reduce attentio…
LFM2.5-24B-A2B is a 24 billion-parameter MoE reasoning model with 2.45 billion active parameters for on-device agentic tasks.
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.