M
MiMo-V2.6 Pro
First among open models at $0.24 per index task, MIT licensed with no revenue threshold
Intelligence v5
30.0
Leaderboard rank
#15
Input / output
$0.35/$1.40
Blended 3:1
$0.61
Output speed
81 tok/s
Value
49 pts/$
Capability profile Normalised to the field maximum
Specifications
Vendor
Xiaomi
Released
2026-09-21
Weights
Open (MIT)
Context
1,000,000 tokens
Max output
Undisclosed
Parameters
1020B total / 42B active
Input price
$0.35 / 1M
Output price
$1.40 / 1M
Cache hit
n/a
Output speed
81 tok/s
Intelligence
30.0（v5）
Benchmark scores
How it is measured →Ranked 15 · bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 Intelligence index | 30.0 | 51% | −29.4 pts |
| AA Intelligence Index v4.3.2 Intelligence index (legacy) | 46.3 | 80% | −11.3 pts |
| GPQA Diamond Graduate-level science reasoning | 59.5 | 62% | −36.9 pts |
| Terminal-Bench 4.0 Terminal-agent tasks | 31.3 | 48% | −33.9 pts |
| LiveBench Design Document generation (DOCX/PPT/PDF) | 78.5 | 90% | −8.4 pts |
| Vals Index Vals composite index | 57.5 | 79% | −15.6 pts |
Assessment
1.02T total parameters with 42B active, 1M context and native text, image, video and audio input. On AA v4.3.2 it ranks first among open models at 46.3, ahead of GLM-5.3 and Kimi K3, and its Vals index of 57.5 is the highest open score. But GPQA Diamond is only 59.5% — science reasoning is a clear weak spot — and the Arena has no human-vote data yet.
At the same price
The four models closest on blended price, side by side.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| DeepSeek V4.1 Flash Open weightsDeepSeek | 32.5 | $0.53 | 148 | +2.5 |
| MiniMax M3.1 FlashMiniMax | 29.0 | $0.26 | 118 | -1.0 |
| GLM-5.3 Open weightsZ.ai | 32.0 | $1.00 | 77 | +2.0 |
| Claude Haiku 5.5Anthropic | 36.2 | $0.20 | 235 | +6.2 |
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard