MiMo V2.6 Flash vs 
GLM 5.3 Flash
Both are SOTA right now. Each row below is one board’s call on the two, and the marked figure is the one that board places higher. No row is added to another, so there is no overall winner: the boards measure different things, and where they split, that is the finding.
| Board | MiMo V2.6 Flash | GLM 5.3 Flash |
|---|---|---|
| Artificial Analysis LLM Leaderboard Artificial Analysis Intelligence Index | 37.9 No. 33 of 372 | 41.8 No. 20 of 372 |
| Arena — Agent Overall | -0.006 No. 34 of 49 | -0.004 No. 33 of 49 |
| Arena — Code Overall | — | 1616 No. 25 of 109 |
| Arena — Text Overall (style control) | 1452 No. 77 of 306 | 1473 No. 41 of 306 |
| Arena — Vision Overall (style control) | 1247 No. 53 of 128 | 1278 No. 30 of 128 |
Each rank is the board’s own: the rank it publishes, or its place in its own order where it publishes none. The bar is how much of that board the model is ahead of. Scores from different boards cannot be compared.
At a glance
| Developer | Xiaomi | Z AI |
|---|---|---|
| Weights | open | open |
| Released | — | 2026-08-26 |
| Price Blended Price, USD per 1M tokens (3:1 input:output) | $0.18 | $0.24 |
| Value: where it sits on the kill line | Executioner | SOTA |
MiMo V2.6 Flash →GLM 5.3 Flash → How the kill line is drawn →
MiMo V2.6 Flash against the other SOTA models
- Qwen 3.8 Flash Next
- Qwen 3.8 Max
- Claude Fable 5
- Claude Fable 5.1
- Claude Opus 5.5
- Claude Sonnet 5.5
- DeepSeek V4.1 Flash
- Gemini 3.8 Flash
- Gemini 4 Argon
- Kimi K3
- Muse Spark 1.3
- GPT 5.6 Terra
- GPT 6.1 Sol
- GPT 6 Astra
- GPT 6 Luna
- Step 5
- Grok 4.7
- MiMo V2.6 Pro
- GLM 5.3