EyesInAI

Switchyard

Benchmark your use caseAI Benchmark Intelligence

Loading models…

EyesInAI·Loading models…

Loading benchmark data…

EyesInAI·Loading live benchmark data

EyesInAI

An independent, measured view of AI models — real cost, real quality, no vendor claims.

Switchyard Benchmark your use case Benchmarks Case studies Task routing Learn News Methodology Contact Privacy

Home Benchmark

Text

Loading models…

EyesInAI·Loading models…

All models

openrouter

mistralai/mistral-small-2603

Last tested Jul 3, 2026

Overall pass

100%

Avg latency

657 ms

Context

262k

Tools

Yes

Input $/1M

$0.15

Output $/1M

$0.60

Tests run

Passed

6/6

AI summary

Hardware to run it

Hosted: $0.15 in / $0.6 out per 1M

Self-host:Single GPU24B (sources vary; some list a 119B MoE)

Q4 ~32GB

~32GB RAM minimum at Q4; ~8 tok/s on an M4 Max

European data sovereignty (GDPR-clean), 256K context, efficient on modest hardware. Param count unsettled across sources (24B vs 119B MoE).

Test results

⚡Ping

Latency & availability — single-word reply

523 ms

🧮Reasoning

Basic math reasoning — show work, give answer

550 ms

{ }JSON Output

Structured output compliance — valid JSON with required keys

895 ms

💻Code Gen

Python function generation with docstring

552 ms

🚀Throughput

Token generation speed — 500-token long-form response

859 ms

🔍Context Recall

Retrieval from in-context data — 20-item list Q&A

Compare with another model See the leaderboard

560 ms