deepseek Models
deepseek logo

V3.1 Terminus (Reasoning)

bydeepseek

Input Price$1.64/1M tokens
Output Price$2.75/1M tokens
Intelligence33.9
Coding33.7

Specifications

Technical details and pricing.

Providerdeepseek
Context WindowN/A
Release DateSep 22, 2025
ModalitiesText

Benchmarks

10 benchmark scores from Artificial Analysis.

GPQA79.2%
MMLU Pro85.1%
HLE15.2%
LiveCodeBench79.8%
AIME 202589.7%
SciCode40.6%
LCR65.0%
IFBench57.0%
Tau237.1%
TerminalBench Hard30.3%

Composite Indices

Higher is better; speed and price are normalized

Standard Benchmarks

Only benchmarks with data are shown

Frequently Asked Questions

What is V3.1 Terminus (Reasoning) good for?

V3.1 Terminus (Reasoning) currently has benchmark and pricing data on this page. Runtime metadata will appear later when the model catalog entry becomes available.

How much does V3.1 Terminus (Reasoning) cost?

Pricing is based on usage. Current rates are $1.64/1M tokens for input and $2.75/1M tokens for output.

Is V3.1 Terminus (Reasoning) available in LLMBase Chat yet?

Not yet on this page. We will add chat and inference actions automatically once an exact catalog entry exists for V3.1 Terminus (Reasoning).

Does V3.1 Terminus (Reasoning) support images or audio?

V3.1 Terminus (Reasoning) focuses on text-based tasks.

Benchmarks and pricing use Artificial Analysis where available. Catalog specs are used as a fallback.