MMODELYST
Models/Cogito v2.1
MDL

Cogito v2.1

Deep Cogito

Verified against Artificial Analysis · Jul 13, 2026
Capability
71
percentile index
Price /M
$1.25 / $1.25
in / out
Context
Avg score
48.9
9 benchmarks
Speed
96 tok/s
TTFT 0.4s
Released
Nov 18, 2025
About

Cogito v2.1 is a proprietary model in the Cogito family from Deep Cogito. Benchmarked on 9 evals, averaging 48.9.

Benchmark scores · 9
avg 48.9 — higher is better
MMLU-Pro
84.9
GPQA Diamond
76.8
AIME
72.7
LiveCodeBench
68.8
IFBench
46.3
SciCode
41
AA Long-Context Reasoning
21.7
Terminal-Bench Hard
16.7
Humanity's Last Exam
11
Details
PublisherDeep Cogito
FamilyCogito
Parameters
License
WeightsClosed
ReleasedNov 18, 2025
Performance & price
Output speed96 tok/s
Latency (TTFT)0.4s
First answer21s
Price in / out$1.25 / $1.25 per 1M
Blended price$1.25 / 1M
Math index72.7

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers
Artificial Analysis$1.25/M