Benched.ai
View
About
Command Palette
Search for a command to run...
Models
Codestral (May '24)
Mistral
Open
Context
Release Date
May 29, 2024
Knowledge Cutoff
Sep 01, 2022
Window
32,768
Pricing
Per 1M tokens
Input
$0.2
Output
$0.6
Blended 3:1
$0.3
Capabilities
Speed
159 t/s
Input
Output
Reasoning tokens
Latency
TTFT
0.29 ms
500 token response
3.43 s
Benchmarks
Intelligence
●○○○○
Math
●○○○○
Coding
●○○○○
MMLU Pro
33.1%
GPQA
25.5%
HLE
5.1%
SciCode
21.8%
AIME
0.0%
MATH 500
35.4%
LiveCodeBench
21.3%
HumanEval
80.2%