- Size
- 235B parameters, 22.0B active (from the repo name)counted from the safetensors header on the Hub
- Licence
- custom, read it, Apache License 2.0Our 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
- 27 Apr 2025the date the repo was created
- Context length
- 40k tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)Qwen3MoeForCausalLM
- On Hugging Face
- Qwen/Qwen3-235B-A22B314k downloads in the last 30 days, 1.1k likes
Scores
- default setting70.7%±2.73 Jun 2025, log
Epoch's name: qwen3-235b-a22b. Best closed model on this set: gpt-6.2-astra_max, 96.8% (OpenAI). All 341 results
- default setting68.9%±0.93 Jun 2025, log
Epoch's name: qwen3-235b-a22b. Best closed model on this set: gpt-5-2025-08-07_high, 98.1% (OpenAI). All 108 results
- Edit format: diff59.6%Aider LLM Leaderboards
Epoch's name: Qwen3-235B-A22B. Best closed model on this set: gpt-5-2025-08-07_high, 88.0% (OpenAI). All 72 results
- default setting75.7%conceptualreasoning.ai
Epoch's name: Qwen3-235B-A22B. Best closed model on this set: claude-fable-5-2_unknown, 99.2% (Anthropic). All 241 results
- default setting67.7%Fiction.live leaderboard
Epoch's name: chutes/Qwen3-235B-A22B. Best closed model on this set: gpt-5-2025-08-07_medium, 97.2% (OpenAI). All 61 results
- default setting83.0%lechmazur/writing Github repository
Epoch's name: qwen3-235b-a22b. Best closed model on this set: gpt-5-2025-08-07_medium, 86.0% (OpenAI). All 49 results
- default setting25.0%20.2 to 29.8conceptualreasoning.ai
Epoch's name: Qwen3-235B-A22B. Best closed model on this set: claude-fable-5-2_xhigh, 70.6% (Anthropic). All 203 results
- default setting31.0%SimpleBench Leaderboard
Epoch's name: qwen3-235b-a22b. Best closed model on this set: claude-fable-5-2_unknown, 94.9% (Anthropic). All 125 results
- default setting37.3%±0.0WeirdML Leaderboard, via Epoch AI
Epoch's name: Qwen3-235B-A22B. Best closed model on this set: gpt-6-astra_promax, 93.6% (OpenAI). All 173 results
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.
Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 1.5 GB, worked out from the model's config (Qwen/Qwen3-235B-A22B), 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
Qwen/Qwen3-235B-A22B-GGUF (the maker's own conversion), 8.5k downloads. Sizes are the real files on the Hub, all parts added up.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| Q4_K_M | 132.4 GBfile size, 5 parts | 134.4 GB132.4 + 1.5 + 0.50 | Commandsllama.cpp llama-server -hf Qwen/Qwen3-235B-A22B-GGUF:Q4_K_MDownload hf download Qwen/Qwen3-235B-A22B-GGUF --include "Q4_K_M/Qwen3-235B-A22B-Q4_K_M-*-of-00005.gguf" --local-dir Qwen3-235B-A22B-GGUF |
| Q5_0 | 150.8 GBfile size, 6 parts | 152.7 GB150.8 + 1.5 + 0.50 | Commandsllama.cpp llama-server -hf Qwen/Qwen3-235B-A22B-GGUF:Q5_0Download hf download Qwen/Qwen3-235B-A22B-GGUF --include "Q5_0/Qwen3-235B-A22B-Q5_0-*-of-00006.gguf" --local-dir Qwen3-235B-A22B-GGUF |
| Q5_K_M | 155.4 GBfile size, 6 parts | 157.3 GB155.4 + 1.5 + 0.50 | Commandsllama.cpp llama-server -hf Qwen/Qwen3-235B-A22B-GGUF:Q5_K_MDownload hf download Qwen/Qwen3-235B-A22B-GGUF --include "Q5_K_M/Qwen3-235B-A22B-Q5_K_M-*-of-00006.gguf" --local-dir Qwen3-235B-A22B-GGUF |
| Q6_K | 179.8 GBfile size, 7 parts | 181.7 GB179.8 + 1.5 + 0.50 | Commandsllama.cpp llama-server -hf Qwen/Qwen3-235B-A22B-GGUF:Q6_KDownload hf download Qwen/Qwen3-235B-A22B-GGUF --include "Q6_K/Qwen3-235B-A22B-Q6_K-*-of-00007.gguf" --local-dir Qwen3-235B-A22B-GGUF |
| Q8_0 | 232.8 GBfile size, 9 parts | 234.7 GB232.8 + 1.5 + 0.50 | Commandsllama.cpp llama-server -hf Qwen/Qwen3-235B-A22B-GGUF:Q8_0Download hf download Qwen/Qwen3-235B-A22B-GGUF --include "Q8_0/Qwen3-235B-A22B-Q8_0-*-of-00009.gguf" --local-dir Qwen3-235B-A22B-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| MLX 4bitmlx-community/Qwen3-235B-A22B-4bit | 123.2 GBfile size | 125.1 GB123.2 + 1.5 + 0.50 | Download page |
| MLX 8bitmlx-community/Qwen3-235B-A22B-8bit | 232.6 GBfile size | 234.6 GB232.6 + 1.5 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| BF16original weights | 437.9 GBfile size | 439.9 GB437.9 + 1.5 + 0.50 |
hf download Qwen/Qwen3-235B-A22B --local-dir Qwen3-235B-A22BOther conversions (17; sizes not read)
- Qwen/Qwen3-235B-A22B-FP8 FP8, by the maker, 6.7k downloads
- Qwen/Qwen3-235B-A22B-GPTQ-Int4 GPTQ, int4, by the maker, 1.4k downloads
- intel-ai/Qwen3-235B-A22B-MXFP4-CT-AutoRound Other, mxfp4, 617 downloads
- QuixiAI/Qwen3-235B-A22B-AWQ AWQ, 316 downloads
- RedHatAI/Qwen3-235B-A22B-NVFP4 NVFP4, 162 downloads
- justinjja/Qwen3-235B-A22B-INT4-W4A16 Other, int4, 159 downloads
- RedHatAI/Qwen3-235B-A22B-FP8-dynamic FP8, 81 downloads
- QuantTrio/Qwen3-235B-A22B-GPTQ-Int8 GPTQ, int8, 60 downloads
- lvkaokao/Qwen3-235B-A22B-MXFP4-Mixed-CT-RTN-AutoRound Other, mxfp4, 38 downloads
- INCModel2/Qwen3-235B-A22B-MXFP8-FP8KV-FP8ATTN-AutoRound Other, 32 downloads
- hyper-accel/Qwen3-235B-A22B-FP8-OC FP8, 25 downloads
- aarnphm/qwen3-235b-a22b-fp8-sharded-tp4 FP8, 23 downloads
- turboderp/Qwen3-235B-A22B-exl3 EXL3, 15 downloads
- async0x42/Qwen3-235B-A22B-exl3_4.0bpw EXL3, 15 downloads
- Minthy/Qwen3-235B-A22B-5.0bpw-h6-exl3 EXL3, 11 downloads
- MikeRoz/Qwen3-235B-A22B-exl3 EXL3, 1 downloads
- abanerjee10/Qwen3-235B-A22B-NVFP4 GPTQ, nvfp4, 0 downloads
The Hub tags 59 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
No provider on Hugging Face Inference Providers serves this model (as of 19 Nov 2026, 13:58 UTC). Other hosts may; we only list what the router reports.
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.
Fine-tunes of this model
Repos that declare this model as their base, most downloaded first. Listing is not a recommendation; unreviewed ones have not been checked.
- nvidia/Qwen3-235B-A22B-NVFP418kUnreviewed
- baichuan-inc/Baichuan-M3-235B235B2.7kUnreviewed
- nvidia/Qwen3-235B-A22B-Eagle3469Unreviewed
- nvidia/Qwen3-235B-A22B-FP8235B337Unreviewed
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.