DeepSeek V3.1 Terminus (Reasoning)

Last updated: Recently

Model Type

Agent

Max Context

164K

Max Output

32,768 tokens

Parameter Scale

DeepSeek V3.1 Terminus is an update to DeepSeek V3.1 that maintains the model& x27;s original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model& x27;s performance in coding and search agents. 163,840 token context window, maximum output of 32,768 tokens. Higher uptime with 4 providers. Includes independent benchmarks from Artificial Analysis.

Input Modality

text

Output Modality

text

Inference Speed

Latest Release Date

9/22/2025

SDK Ecosystem

artificial-analysismanufactureraa-bootstrap

Model Overview Fields

name

DeepSeek V3.1 Terminus (Reasoning)

Release date

9/22/2025

Performance

model_creator.name

DeepSeek

AA Evaluation Scores

AA Intelligence Index

33.9

AA Coding Index

33.7

AA Math Index

89.7

MMLU Pro

0.851

GPQA

0.792

HLE

0.152

LiveCodeBench

0.798

SciCode

0.406

Math-500

AIME

AIME 2025

0.896666666666667

IFBench

0.570068027210884

LCR

0.65

TerminalBench Hard

0.303030303030303

TAU2

0.371345029239766

Output Speed (tokens/s)

TTFT (s)

First Answer Token (s)

DeepSeek

DeepSeek V3.1 Terminus (Reasoning)

$1.64/M$2.75/M

There are no reviews for this model yet.

artificial-analysismanufactureraa-bootstrap

0.0