modelgrep

Anthropic: Claude Opus 4.8 vs Xiaomi: MiMo-V2.5-Pro

Anthropic: Claude Opus 4.8 wins on more metrics (5 of 9), but the right pick depends on what you optimize for — see the breakdown below.

MetricAnthropic: Claude Opus 4.8Xiaomi: MiMo-V2.5-Pro
Intelligence Index61.453.8
Coding Index56.745.5
GPQA Diamond92%87%
Design Arena Elo13721334
Speed (tokens/sec)5938
Latency1.8s255ms
Input price /M$5.00$0.435
Output price /M$25.00$0.870
Context window1M1.0M
CapabilitiesReasoningToolsJSONVisionReasoningToolsJSON