MiMo-V2.5-Pro mimo-v2.5-pro
xiaomi · mimo reasoning flagship
#5 overall
- params
- 1023B
- arch
- moe
- context
- 1024k
- license
- mit open weights
- released
- Apr 2026
- reasoning
- yes
- train compute
- 6.8e+24 FLOP
- downloads/30d
- 101.7k
# why ranked
Overall: #5 — score 49.1 ● high — all signals present
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.50 | 42.2 |
| arena_elo | 0.30 | 49.3 |
| bench_composite | 0.20 | 66.3 |
Coding: provisional — score 89.7 ● low — a single signal; treat with caution
provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.
| signal | weight | input (0–100) |
|---|---|---|
| aa_coding | 0.25 | 89.7 |
| aider_polyglot | 0.25 | — |
| arena_elo_code | 0.15 | — |
| swe_bench_verified | 0.35 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
collecting history — trends appear after a few daily runs.
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Coding Index | 60.2 | Artificial Analysis | — |
| AA Intelligence Index | 42.2 | Artificial Analysis | — |
| LMArena Elo | 1466.0 | LMArena | Jul 2026 |
| Aime | 0.4 / 1 | LLM Stats | — |
| Arc-c | 1.0 / 1 | LLM Stats | — |
| Bbh | 0.9 / 1 | LLM Stats | — |
| C-eval | 0.9 / 1 | LLM Stats | — |
| Claw-eval | 0.6 / 1 | LLM Stats | — |
| Cmmlu | 0.9 / 1 | LLM Stats | — |
| Drop | 0.9 / 1 | LLM Stats | — |
| Frontier-swe-impl | 3.4 / 1 | LLM Stats | — |
| Gdpval-aa | 1286.0 / 3000 | LLM Stats | — |
| Gsm8k | 1.0 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M blend | tps | uptime |
|---|---|---|---|---|---|---|---|
| AtlasCloud | fp8 | 1000k | $0.43 | $0.87 | $0.54 | — | 97.3% |
| Xiaomi | fp8 | 1024k | $0.43 | $0.87 | $0.54 | — | 99.7% |
| Novita | unknown | 1024k | $0.52 | $1.04 | $0.65 | — | 99.8% |
| DigitalOcean | unknown | 85k | $0.60 | $3.00 | $1.20 | — | 98.8% |
| DeepInfra | fp8 | 1024k | $1.00 | $3.00 | $1.50 | — | 99.5% |
blended = (3·input + output) / 4 per 1M tokens · cheapest first
source aliases
- arena
mimo-v2.5-pro- epoch
MiMo-V2.5-Pro- llmstats
mimo-v2.5-pro- openrouter
xiaomi/mimo-v2.5-pro