- Size
- 2.4T parameters, 95.0B active (from the repo name)counted from the safetensors header on the Hub
- Licence
- custom, read it, Qwen3.8-Max LicenseOur reading, check the licence: A custom licence we have not read for this model: other makers use the same name for different texts. Read the model's own licence file before you use it.Licence text
- On the Hub since
- 8 Aug 2026the date the repo was created
- Context length
- 256k tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)Qwen3_5MoeForCausalLM
- On Hugging Face
- Qwen/Qwen3.8-2.4T-A95B30k downloads in the last 30 days, 1.4k likes
Scores
No independent scores for this model in Epoch AI's data. A score is only shown here when Epoch's name for a model matches this repo with certainty; an uncertain match is left out rather than guessed.
Claimed by the maker Claimed
The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.
Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.72 GB, worked out from the model's config (Qwen/Qwen3.8-2.4T-A95B), at 16-bit precision. Choose a machine above to see a verdict for each format. Method
GGUF for llama.cpp, Ollama, LM Studio and others
unsloth/Qwen3.8-2.4T-A95B-GGUF (converted by unsloth), 11k downloads. Sizes are the real files on the Hub, all parts added up.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| UD-Q1_0 | 370.0 GBfile size, 10 parts | 371.2 GB370.0 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-Q1_0Download hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-Q1_0/Qwen3.8-2.4T-A95B-UD-Q1_0-*-of-00010.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| UD-IQ1_S | 473.5 GBfile size, 12 parts | 474.7 GB473.5 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-IQ1_SDownload hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-IQ1_S/Qwen3.8-2.4T-A95B-UD-IQ1_S-*-of-00012.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| UD-IQ1_M | 525.2 GBfile size, 13 parts | 526.4 GB525.2 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-IQ1_MDownload hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-IQ1_M/Qwen3.8-2.4T-A95B-UD-IQ1_M-*-of-00013.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| UD-IQ2_XXS | 611.5 GBfile size, 15 parts | 612.7 GB611.5 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-IQ2_XXSDownload hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-IQ2_XXS/Qwen3.8-2.4T-A95B-UD-IQ2_XXS-*-of-00015.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| UD-IQ2_XS | 680.5 GBfile size, 16 parts | 681.7 GB680.5 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-IQ2_XSDownload hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-IQ2_XS/Qwen3.8-2.4T-A95B-UD-IQ2_XS-*-of-00016.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| UD-IQ3_XXS | 889.9 GBfile size, 21 parts | 891.1 GB889.9 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-IQ3_XXSDownload hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-IQ3_XXS/Qwen3.8-2.4T-A95B-UD-IQ3_XXS-*-of-00021.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| UD-IQ4_XS | 1220.8 GBfile size, 29 parts | 1222.1 GB1220.8 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:UD-IQ4_XSDownload hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "UD-IQ4_XS/Qwen3.8-2.4T-A95B-UD-IQ4_XS-*-of-00029.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| Q8_0 | 2421.7 GBfile size, 56 parts | 2422.9 GB2421.7 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:Q8_0Download hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "Q8_0/Qwen3.8-2.4T-A95B-Q8_0-*-of-00056.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
| BF16 | 4557.1 GBfile size, 140 parts | 4558.3 GB4557.1 + 0.72 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/Qwen3.8-2.4T-A95B-GGUF:BF16Download hf download unsloth/Qwen3.8-2.4T-A95B-GGUF --include "BF16/Qwen3.8-2.4T-A95B-BF16-*-of-00140.gguf" --local-dir Qwen3.8-2.4T-A95B-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| MLX 4bitpipenetwork/Qwen3.8-2.4T-A95B-MLX-reap75-4bit | 336.1 GBfile size | 337.4 GB336.1 + 0.72 + 0.50 | Download page |
| MLX 3bitkernelpool/Qwen3.8-2.4T-A95B-3bit-UVMAX | 750.3 GBfile size | 751.5 GB750.3 + 0.72 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| BF16original weights | 4556.4 GBfile size | 4557.6 GB4556.4 + 0.72 + 0.50 |
hf download Qwen/Qwen3.8-2.4T-A95B --local-dir Qwen3.8-2.4T-A95BOther conversions (11; sizes not read)
- Qwen/Qwen3.8-2.4T-A95B-FP8 FP8, by the maker, 11k downloads
- nvidia/Qwen3.8-2.4T-A95B-NVFP4 NVFP4, 2.4k downloads
- Inferact/Qwen3.8-2.4T-A95B-NVFP4 NVFP4, 593 downloads
- RadixArk/Qwen3.8-2.4T-A95B-NVFP4 NVFP4, 351 downloads
- RedHatAI/Qwen3.8-2.4T-A95B-NVFP4-REAP-25 NVFP4, 322 downloads
- a4lg/Qwen3.8-2.4T-A95B-MTP-ONLY-GGUF GGUF, 238 downloads
- Inferact/Qwen3.8-2.4T-A95B-MXFP4 Other, mxfp4, 184 downloads
- RedHatAI/Qwen3.8-2.4T-A95B-FP8 FP8, 132 downloads
- Xihnik/Qwen3.8-2.4T-A95B-FP8 FP8, 58 downloads
- LilacGhost10/Qwen3.8-2.4T-A95B-FP8 FP8, 51 downloads
- dudeman2512/Qwen3.8-2.4T-A95B-int4 Other, int4, 33 downloads
The Hub tags 34 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
4 providers serve this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.
| Provider | In | Out | Context | Precision | Tokens/s | First token | Uptime 24h | Tools | JSON |
|---|---|---|---|---|---|---|---|---|---|
| deepinfra | $2.00 | $6.00 | 256k | not given | 69 | 0.7 s | ? | yes | yes |
| Novita | $2.00 | $6.00 | 977k | not given | 45 | 1.3 s | 99.7% | yes | yes |
| SiliconFlow | $2.00 | $6.00 | 1M | fp8 | 33 | 1.4 s | 99.5% | yes | yes |
| Venice | $2.00 | $6.00 | 256k | not given | 123 | 2.5 s | 99.7% | yes | no |
Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.
curl https://router.huggingface.co/v1/chat/completions \
-H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \
-d '{"model": "Qwen/Qwen3.8-2.4T-A95B:deepinfra", "messages": [{"role": "user", "content": "Hello"}]}'Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.
Planned for this page Not built yet
- Try it in private chat
- A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
- Fine-tune it
- Fine-tuning a few base models from this page and publishing the result here.
- Creators and earnings
- Pages for people who publish fine-tunes and adapters of this model, and what they earn.
- Fund and bounties
- Bounties for new open models, paid from a public fund. How the fund will work.
- Contribute compute
- How to give compute to open model work, with Akash.
Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.