MiniMax-M2.5
- In Input modalities: Text Out Output modalities: Text
- Reasoning
- Tool use
- Open weights
Prices
2 offerings · prices as listedMiniMax-M2.5| Conditions/ 1M tokens | Input | Cache read | Cache write | Output |
|---|---|---|---|---|
| Standard no longer listed since Oct 19, 2026 | $0.30 | $0.030 | $0.375 | $1.20 |
Prices from MiniMax pay-as-you-go pricing (opens in a new tab) (checked Oct 19, 2026)
MiniMax-M2.5| Conditions/ 1M tokens | Input | Cache read | Cache write | Output |
|---|---|---|---|---|
| Standard no longer listed since Oct 19, 2026 | ¥2.10 | ¥0.21 | ¥2.625 | ¥8.40 |
Prices from MiniMax 按量计费价格 (opens in a new tab) (checked Oct 18, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Licence
- Modified-MIT
- Family
- minimax-m
- Model ID
Not stated on MiniMax’s pages: max input, reasoning efforts, reasoning can be off, max thinking budget, structured output, built-in tools, knowledge cutoff, aliases.
Run locally
Weights: MiniMaxAI/MiniMax-M2.5- Parameters 228.7B
- Active 11B (estimated) · 256 experts, 8 per token
- Native precision FP8
- Layers 62
- KV cache per token 248 KB
- Config positions 192K (config.json max_position_embeddings)
- API context 200K (maker’s API)
- Needs
- ≈ 120.7 GB (weights 113.1 + KV cache 1.9 + overhead 5.7)
- Usable
- 21.6 GB (90% of its memory)
- Try
- IQ2_XXS runs (needs system memory), but below 4 bits quality drops noticeably
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| FP8 (native)maker’s file | 214.3 GB | ≈ 227.0 | ✕Doesn’t fit | — |
| Q6_K | 174.9 GB | ≈ 185.6 | ✕Doesn’t fit | — |
| Q5_K_M | 151.5 GB | ≈ 161.0 | ✕Doesn’t fit | — |
| Q5_0 | 146.4 GB | ≈ 155.7 | ✕Doesn’t fit | — |
| Q4_K_M | 129.4 GB | ≈ 137.8 | ✕Doesn’t fit | — |
| Q4_0 | 119.8 GB | ≈ 127.7 | ✕Doesn’t fit | — |
| IQ4_XS | 113.1 GB | ≈ 120.7 | ✕Doesn’t fit | — |
| Q3_K_Mquality drops | 104.1 GB | ≈ 111.2 | ✕Doesn’t fit | — |
| IQ3_XXSquality drops | 81.5 GB | ≈ 87.5 | ✕Doesn’t fit | — |
| Q2_Kquality drops | 79.9 GB | ≈ 85.8 | ✕Doesn’t fit | — |
| IQ2_XXSquality drops | 54.8 GB | ≈ 59.5 | ○Needs system memory 24–35 tok/s | — |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): GPU NVIDIA B200 (HGX B200); unified memory Apple M5 Ultra (64-core GPU) · 256 GB.
Run commands
The maker publishes no files at this precision. Choose its native precision or one of its own builds, or quantise it yourself with llama.cpp.
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated
Cited scores
From third-party leaderboards, not measured by us- LMArena Text 1,391
#108 of 163 listed models · CI 1,387–1,395 · 42,783 votes
- EQ-Bench Creative Writing v3 1,362
#70 of 93 listed models
- EQ-Bench Longform Writing 65.3
#50 of 92 listed models
- LMArena Text · Chinese 1,430
#99 of 161 listed models · CI 1,419–1,442 · 2,871 votes
- LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 28, 2026
- EQ-Bench by Samuel J. Paech (eqbench.com; EQ-bench/EQ-bench-site on GitHub), MIT License. · License · as of Oct 28, 2026
- changes we made
Included in coding plans
- Model Studio Coding Plan Pro $50.00 / month6,000 requests every 5 hours45,000 requests per week90,000 requests per month
- Kiro Free Free50 credits per month
- Kiro Pro $20.00 / month1,000 credits per month
- Kiro Pro+ $40.00 / month2,000 credits per month
- Kiro Pro Max $100.00 / month5,000 credits per month
- Kiro Power $200.00 / month10,000 credits per month
Changes
All changes →No changes recorded yet.
Sources
- MiniMax pay-as-you-go pricing (opens in a new tab) platform.minimax.io · checked Oct 19, 2026
- MiniMax 按量计费价格in Chinese (opens in a new tab) platform.minimaxi.com · checked Oct 18, 2026
- MiniMax text generation (opens in a new tab) platform.minimax.io · checked Oct 19, 2026
- Anthropic SDK (opens in a new tab) platform.minimax.io · checked Nov 1, 2026
- MiniMax model release notes (opens in a new tab) platform.minimax.io · checked Nov 1, 2026
- MiniMaxAI/MiniMax-M2.5 model card (opens in a new tab) huggingface.co · checked Nov 2, 2026
- models.dev: MiniMax-M2.5 (community data, MIT) (opens in a new tab) github.com · checked Sep 30, 2026
- MiniMaxAI/MiniMax-M2.5 model card (opens in a new tab) huggingface.co · checked Nov 5, 2026
- MiniMaxAI/MiniMax-M2.5 model card (opens in a new tab) huggingface.co · checked Nov 4, 2026
- MiniMaxAI/MiniMax-M2.5 config.json (opens in a new tab) huggingface.co · checked Nov 5, 2026
- MiniMaxAI/MiniMax-M2.5 files (opens in a new tab) huggingface.co · checked Oct 31, 2026
MiniMax’s official policies: data use, privacy, terms →
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
MiniMax · China · minimax.io
Related models
Snapshots, the same family and other models from this provider