Back

DeepSeek

DeepSeek V3.1 Terminus

ReasoningOpen weights
Intelligence
14.8
Rank 313 of 744 measured models
Coding
43.5
Rank 146 of 298 measured models
Price/1M
$1.91
Rank 458 of 659 measured models
Context window
128K
Rank 773 of 1199 measured models

DeepSeek V3.1 Terminus (Reasoning) scores 14.8 on the Artificial Analysis Intelligence Index, ranking 313th of 744 measured models. The closest models on this index are Gemini 2.5 Pro Preview (Mar' 25) (15.0), GLM-4.6 (14.9), and Gemma 4 31B (14.7). At $1.91 per million tokens (3:1 input/output blended price), it is more expensive than GLM-4.6 ($0.981).

Artificial Analysis Indices
Reasoning
Intelligence14.8
Rank 313 of 744 measured models
Coding43.5
Rank 146 of 298 measured models
Math89.7
Rank 26 of 270 measured models
Agentic6.8
Rank 127 of 189 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Price / 1M tokens
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting15.3
Rank 168 of 218 measured models
Strategy & Ops11.4
Rank 170 of 218 measured models
Legal18.1
Rank 166 of 218 measured models
Engineering18.1
Rank 170 of 217 measured models
Economics23.1
Rank 174 of 225 measured models
Standard Benchmarks
MMLU Pro85.1%
Rank 30 of 349 measured models
GPQA Diamond79.2%
Rank 213 of 619 measured models
HLE16.4%
Rank 249 of 664 measured models
LiveCodeBench79.8%
Rank 29 of 343 measured models
SciCode38.0%
Rank 188 of 231 measured models
AIME 202589.7%
Rank 26 of 270 measured models
IFBench57.0%
Rank 143 of 450 measured models
AA-LCR69.3%
Rank 228 of 575 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy