Intelligence Index
0122.3
> 57% OF MODELS
MiMo V2.5
Updated: September 2026
Native omnimodal MiMo with 1M context, strong multimodal perception, and Pro-level agentic performance at half the cost.
Specifications
- Context
- 1M
- Input Modalities
- Image, Text, File
- Output Modalities
- Text
- Input
- $0.4/M
- Output
- $2/M
Capabilities
VISIONTHINKINGCODING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Qwen3.7 Plus | Qwen | 1M | $0.4/M | $1.6/M |
| Qwen3.5 Plus | Qwen | 1M | $0.4/M | $2.4/M |
| Devstral 2 2512 | MistralAI | 262K | $0.4/M | $2/M |
| Kimi K2 | Moonshot | 131K | $0.4/M | $2/M |
| Qwen3 VL 235B A22B Thinking | Qwen | 131K | $0.4/M | $4/M |
Performance Metrics
Coding Index
0256.8
> 65% OF MODELS
Agentic Index
0317.4
> 45% OF MODELS
Average Response Performance
- Output Speed
- 45.2 tok/s
- Time To First Token
- 6.91s
- Time To First Answer Token
- 51.13s
- End To End Response Time
- 62.19s
DATA SOURCE: Artificial Analysis