MiMo-V2.6-Pro
The strongest open-weight model right now and absurdly cheap for what it does. MIT-licensed 1T MoE that lands within a few points of Muse Spark and Grok 4.7. Slow-ish and verbose, so budget for latency and thinking tokens.
Key facts
- Score
- 81 / 100, rank #7Evals81Price66People78Fit100
- Price
- $0.43 in · $0.87 out per 1M tokens (MiMo-V2.6-Pro - Artificial Analysis)
- Blended price
- $0.54 per 1M tokens (3 input : 1 output)
- Context window
- 1M
- Open weights
- Yes
- Released
- Sep 21, 2026
- Best for
- Self-hosted agentic coding, Cheap frontier-class batch reasoning, Replacing closed mid-tier models
MiMo-V2.6-Pro costs $0.43 per million input tokens and $0.87 per million output tokens on Xiaomi's list price (MiMo-V2.6-Pro - Artificial Analysis, read Oct 11, 2026).
Independent evals
| Benchmark | Result | Who ran it | Read |
|---|---|---|---|
| Artificial Analysis Intelligence Index | 46 | MiMo-V2.6-Pro - Artificial Analysis | Oct 10, 2026 |
| Arena Text score | 1480 (rank 26) | Text Arena Leaderboard | Oct 10, 2026 |
| Arena WebDev score | 1629 (rank 19) | WebDev Arena Leaderboard | Oct 10, 2026 |
- Artificial Analysis output speed (tokens/s): 47.5 (MiMo-V2.6-Pro - Artificial Analysis)
What practitioners say
HN (1,130 points) was won over less by scores than by Xiaomi's transparency: a live training dashboard with loss curves, dropped datasets and running cost estimates. Users call it 'incredibly cheap' for Muse-Spark-level benchmarks and say it keeps DeepSeek and GLM 'in check'. Complaints are consistent: 'faster than DeepSeek but still much slower than leading models', and it 'seems to overthink way too much', producing huge reasoning traces for trivial edits. Several recommend it via OpenRouter rather than the Chinese first-party API.
Caveats
- Overthinks: long, repetitive reasoning traces on simple edits; AA flags it as verbose vs the median model.
- ~47 tok/s on the first-party API is slow next to Gemini Flash or Haiku-class models.
- Only 4k Arena votes so far; the rank is still settling.
Questions people ask
How much does MiMo-V2.6-Pro cost?
MiMo-V2.6-Pro costs $0.43 per million input tokens and $0.87 per million output tokens on Xiaomi's list price (MiMo-V2.6-Pro - Artificial Analysis, read Oct 11, 2026).
How good is MiMo-V2.6-Pro?
MiMo-V2.6-Pro ranks #7 of 26 on the Models at Work leaderboard with a score of 81 out of 100 (October 2026 edition). The strongest open-weight model right now and absurdly cheap for what it does. MIT-licensed 1T MoE that lands within a few points of Muse Spark and Grok 4.7. Slow-ish and verbose, so budget for latency and thinking tokens.
What is MiMo-V2.6-Pro best for?
MiMo-V2.6-Pro is best for self-hosted agentic coding, cheap frontier-class batch reasoning, replacing closed mid-tier models.
What are MiMo-V2.6-Pro's benchmark scores?
Artificial Analysis Intelligence Index: 46 (MiMo-V2.6-Pro - Artificial Analysis); Arena Text score: 1480 (rank 26) (Text Arena Leaderboard); Arena WebDev score: 1629 (rank 19) (WebDev Arena Leaderboard).
What is MiMo-V2.6-Pro's context window?
MiMo-V2.6-Pro has a 1M token context window according to Xiaomi.
Is MiMo-V2.6-Pro open weights?
Yes. MiMo-V2.6-Pro is released with open weights, so it can be self-hosted.
Compare MiMo-V2.6-Pro
Methodology: Score is 0 to 100 and computed, not typed: 55% independent evals (Artificial Analysis, Arena, Scale SEAL, Epoch, each scaled against its natural floor and the best score in this table, then averaged), 15% blended price on a fixed log scale ($0.05 per million is 100, $60 is 0), 15% practitioner sentiment from Signals and community threads, 15% operational fit (context window, open weights). A missing component drops out and the rest are reweighted. A model with no independent eval yet is provisional and ranks below every measured one. Within three points is a tie. Vendor-published figures are labelled and never counted.