Intelligence Index
0130.4
> 51% OF MODELS
MistralAI
Mistral Medium 3.5
Updated: August 2026
Mistral Medium 3.5 is built for agentic workflows, coding, and long-horizon reasoning. It supports text and image inputs with a 256K context window and reliable multi-tool calling.
Specifications
- Context
- 262K
- Input Modalities
- Text, Image, File
- Output Modalities
- Text
- Input
- $1.5/M
- Output
- $7.5/M
Capabilities
VISIONTHINKINGCODING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| Gemini 3.6 Flash | Google | 1M | $1.5/M | $7.5/M |
| Gemini 3.6 Flash | Google | 1M | $1.5/M | $7.5/M |
| Gemini 3.5 Flash | Google | 1M | $1.5/M | $9/M |
| Gemini 3.5 Flash | Google | 1M | $1.5/M | $9/M |
| GPT-3.5 Turbo Instruct | OpenAI | 4K | $1.5/M | $2/M |
Performance Metrics
Coding Index
0246.9
> 52% OF MODELS
Agentic Index
0319.2
> 39% OF MODELS
Average Response Performance
- Output Speed
- 131.9 tok/s
- Time To First Token
- 2.28s
- Time To First Answer Token
- 17.44s
- End To End Response Time
- 21.23s
DATA SOURCE: Artificial Analysis