- Size
- 229B parameterscounted from the safetensors header on the Hub
- Licence
- custom, read it, MiniMax Model 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
- 12 Feb 2026the date the repo was created
- Context length
- 192k tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)MiniMaxM2ForCausalLM
- On Hugging Face
- MiniMaxAI/MiniMax-M2.5967k downloads in the last 30 days, 1.5k likes
Scores
- default setting63.7%via Epoch AI
Epoch's name: MiniMax-M2.5. Best closed model on this set: claude-fable-5_max, 98.5% (Anthropic). All 318 results
- default setting4.9%via Epoch AI
Epoch's name: MiniMax-M2.5. Best closed model on this set: gpt-6-astra_max, 95.0% (OpenAI). All 299 results
- default setting11.4%via Epoch AI
Epoch's name: MiniMax-M2.5. Best closed model on this set: gpt-5.4-2026-03-05_xhigh, 27.9% (OpenAI). All 23 results
- default setting6.3%via Epoch AI
Epoch's name: MiniMax-M2.5. Best closed model on this set: gpt-5.5_high, 22.2% (OpenAI). All 17 results
- default setting4.0%±2.0vals.ai
Epoch's name: MiniMax-M2.5. Best closed model on this set: claude-fable-5-1_max, 100.0% (Anthropic). All 85 results
Epoch's name: MiniMax-M2.5. Best closed model on this set: gpt-5.5_unknown, 84.7% (OpenAI). All 202 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: 1.9 GB, worked out from the model's config (MiniMaxAI/MiniMax-M2.5), 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/MiniMax-M2.5-GGUF (converted by unsloth), 5.6k downloads. Sizes are the real files on the Hub, all parts added up.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| UD-IQ3_XXS | 86.9 GBfile size, 3 parts | 89.3 GB86.9 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:UD-IQ3_XXSDownload hf download unsloth/MiniMax-M2.5-GGUF --include "UD-IQ3_XXS/MiniMax-M2.5-UD-IQ3_XXS-*-of-00003.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q3_K_S | 91.9 GBfile size, 3 parts | 94.4 GB91.9 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q3_K_SDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q3_K_S/MiniMax-M2.5-Q3_K_S-*-of-00003.gguf" --local-dir MiniMax-M2.5-GGUF |
| UD-Q3_K_XL | 94.3 GBfile size, 4 parts | 96.8 GB94.3 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:UD-Q3_K_XLDownload hf download unsloth/MiniMax-M2.5-GGUF --include "UD-Q3_K_XL/MiniMax-M2.5-UD-Q3_K_XL-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q3_K_M | 101.8 GBfile size, 4 parts | 104.2 GB101.8 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q3_K_MDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q3_K_M/MiniMax-M2.5-Q3_K_M-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| IQ4_XS | 113.5 GBfile size, 4 parts | 116.0 GB113.5 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:IQ4_XSDownload hf download unsloth/MiniMax-M2.5-GGUF --include "IQ4_XS/MiniMax-M2.5-IQ4_XS-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| MXFP4_MOE | 115.3 GBfile size, 4 parts | 117.7 GB115.3 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:MXFP4_MOEDownload hf download unsloth/MiniMax-M2.5-GGUF --include "MXFP4_MOE/MiniMax-M2.5-MXFP4_MOE-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| IQ4_NL | 120.1 GBfile size, 4 parts | 122.6 GB120.1 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:IQ4_NLDownload hf download unsloth/MiniMax-M2.5-GGUF --include "IQ4_NL/MiniMax-M2.5-IQ4_NL-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q4_0 | 120.6 GBfile size, 4 parts | 123.0 GB120.6 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q4_0Download hf download unsloth/MiniMax-M2.5-GGUF --include "Q4_0/MiniMax-M2.5-Q4_0-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q4_K_S | 121.1 GBfile size, 4 parts | 123.5 GB121.1 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q4_K_SDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q4_K_S/MiniMax-M2.5-Q4_K_S-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| UD-Q4_K_XL | 122.3 GBfile size, 4 parts | 124.7 GB122.3 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:UD-Q4_K_XLDownload hf download unsloth/MiniMax-M2.5-GGUF --include "UD-Q4_K_XL/MiniMax-M2.5-UD-Q4_K_XL-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q4_K_M | 128.8 GBfile size, 4 parts | 131.3 GB128.8 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q4_K_MDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q4_K_M/MiniMax-M2.5-Q4_K_M-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q4_1 | 133.4 GBfile size, 4 parts | 135.8 GB133.4 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q4_1Download hf download unsloth/MiniMax-M2.5-GGUF --include "Q4_1/MiniMax-M2.5-Q4_1-*-of-00004.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q5_K_S | 146.7 GBfile size, 5 parts | 149.1 GB146.7 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q5_K_SDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q5_K_S/MiniMax-M2.5-Q5_K_S-*-of-00005.gguf" --local-dir MiniMax-M2.5-GGUF |
| UD-Q5_K_XL | 150.7 GBfile size, 5 parts | 153.1 GB150.7 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:UD-Q5_K_XLDownload hf download unsloth/MiniMax-M2.5-GGUF --include "UD-Q5_K_XL/MiniMax-M2.5-UD-Q5_K_XL-*-of-00005.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q5_K_M | 151.2 GBfile size, 5 parts | 153.6 GB151.2 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q5_K_MDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q5_K_M/MiniMax-M2.5-Q5_K_M-*-of-00005.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q6_K | 174.9 GBfile size, 5 parts | 177.3 GB174.9 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q6_KDownload hf download unsloth/MiniMax-M2.5-GGUF --include "Q6_K/MiniMax-M2.5-Q6_K-*-of-00005.gguf" --local-dir MiniMax-M2.5-GGUF |
| UD-Q6_K_XL | 180.9 GBfile size, 5 parts | 183.4 GB180.9 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:UD-Q6_K_XLDownload hf download unsloth/MiniMax-M2.5-GGUF --include "UD-Q6_K_XL/MiniMax-M2.5-UD-Q6_K_XL-*-of-00005.gguf" --local-dir MiniMax-M2.5-GGUF |
| Q8_0 | 226.4 GBfile size, 6 parts | 228.9 GB226.4 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:Q8_0Download hf download unsloth/MiniMax-M2.5-GGUF --include "Q8_0/MiniMax-M2.5-Q8_0-*-of-00006.gguf" --local-dir MiniMax-M2.5-GGUF |
| UD-Q8_K_XL | 243.4 GBfile size, 7 parts | 245.9 GB243.4 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:UD-Q8_K_XLDownload hf download unsloth/MiniMax-M2.5-GGUF --include "UD-Q8_K_XL/MiniMax-M2.5-UD-Q8_K_XL-*-of-00007.gguf" --local-dir MiniMax-M2.5-GGUF |
| BF16 | 426.1 GBfile size, 10 parts | 428.5 GB426.1 + 1.9 + 0.50 | Commandsllama.cpp llama-server -hf unsloth/MiniMax-M2.5-GGUF:BF16Download hf download unsloth/MiniMax-M2.5-GGUF --include "BF16/MiniMax-M2.5-BF16-*-of-00010.gguf" --local-dir MiniMax-M2.5-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| MLX 6bitlmstudio-community/MiniMax-M2.5-MLX-6bit | 173.1 GBfile size | 175.5 GB173.1 + 1.9 + 0.50 | Download page |
| MLX 8bitlmstudio-community/MiniMax-M2.5-MLX-8bit | 226.3 GBfile size | 228.7 GB226.3 + 1.9 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| F8_E4M3original weights | 214.3 GBfile size | 216.7 GB214.3 + 1.9 + 0.50 |
hf download MiniMaxAI/MiniMax-M2.5 --local-dir MiniMax-M2.5Other conversions (19; sizes not read)
- nvidia/MiniMax-M2.5-NVFP4 NVFP4, 31k downloads
- mudler/MiniMax-M2.5-APEX-GGUF GGUF, 1.6k downloads
- amd/MiniMax-M2.5-MXFP4 Other, mxfp4, 607 downloads
- mratsim/MiniMax-M2.5-FP8-INT4-AWQ AWQ, fp8, 351 downloads
- RedHatAI/MiniMax-M2.5-NVFP4 NVFP4, 249 downloads
- cyankiwi/MiniMax-M2.5-AWQ-4bit AWQ, 4bit, 232 downloads
- Intel/MiniMax-M2.5-int4-AutoRound Other, int4, 127 downloads
- unsloth/MiniMax-M2.5 FP8, 119 downloads
- QuantTrio/MiniMax-M2.5-AWQ AWQ, 106 downloads
- cerebras/MiniMax-M2.5-REAP-139B-A10B FP8, 105 downloads
- cerebras/MiniMax-M2.5-REAP-172B-A10B FP8, 99 downloads
- lukealonso/MiniMax-M2.5-NVFP4 NVFP4, 93 downloads
- RedHatAI/MiniMax-M2.5-quantized.w4a16 Other, w4a16, 89 downloads
- mratsim/MiniMax-M2.5-BF16-INT4-AWQ AWQ, bf16, 80 downloads
- RedHatAI/MiniMax-M2.5-quantized.w8a8 Other, w8a8, 63 downloads
- amd/MiniMax-M2.5-NVFP4 NVFP4, 58 downloads
- chenzs/MiniMax-M2.5-catid-FP8-Serve FP8, 44 downloads
- MikeRoz/MiniMax-M2.5-exl3 EXL3, 3 downloads
- NeuroSenko/MiniMax-M2.5-exl3 EXL3, 3 downloads
The Hub tags 64 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 |
|---|---|---|---|---|---|---|---|---|---|
| Venice | $0.27 | $0.95 | 193k | not given | 58 | 0.9 s | 98.2% | yes | no |
| DigitalOcean | $0.30 | $1.20 | 64k | not given | 71 | 0.6 s | 81.7% | yes | yes |
| novita | $0.30 | $1.20 | 200k | not given | 46 | 0.8 s | ? | yes | no |
| featherless-ai | ? | ? | ? | not given | ? | ? | ? | not given | not given |
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": "MiniMaxAI/MiniMax-M2.5:venice", "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.