ModelBeat
Models/Cogito v2.1
All models
D

Deep Cogito: Cogito v2.1

Watch this modelCompareFind source
cogito-v2-1

Cogito v2.1 is Deep Cogito’s AI model, released Nov 19, 2025.

Context
—
Input
— / 1M
Output
— / 1M
Released
Nov 19, 2025
Modalities

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Cogito v2.1 lands among all models tracked on Model Beat.

26.0
Intelligence Index
Artificial Analysis
26th percentile of tracked models
32.0
Coding Index
Artificial Analysis
32nd percentile of tracked models

Reasoning

3 evals
GPQA Diamond76.8%

Graduate-level scientific reasoning

Humanity's Last Exam11.0%

Frontier of human expert knowledge

MMLU-Pro84.9%

Broad expert knowledge

Coding

2 evals
LiveCodeBench68.8%

Contamination-free coding

SciCode41.0%

Scientific research coding

Math

1 evals
AIME 2024/202572.7%

Olympiad-qualifier math

Compare Cogito v2.1 with

Cogito v2.1 vs GPT-5.6 SolCogito v2.1 vs Claude Fable 5Cogito v2.1 vs Gemini 3.1 ProCogito v2.1 vs Muse SparkCogito v2.1 vs Grok 4.5

Frequently asked questions

What is Cogito v2.1?

Cogito v2.1 is an AI model, from Deep Cogito, released Nov 19, 2025.

Who created Cogito v2.1?

Cogito v2.1 is an AI model developed by Deep Cogito, released Nov 19, 2025.

How does Cogito v2.1 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Cogito v2.1 scores 84.9% on MMLU-Pro, 41.0% on SciCode, 76.8% on GPQA Diamond.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Jul 30, 2026.