modelgrep
D

DeepSeek: DeepSeek V3.1 Terminus

deepseek/deepseek-v3.1-terminus

82nd smartest of 180ReasoningToolsJSON
Use via OpenRouter ↗
Intelligence
28.5
82nd of 180
Design Elo
1238
UI Component
Speed
26
254th fastest
Latency
1.1s
first token
Input price
$0.270
124th cheapest
Context
164K
33K max out

How it compares

Smarter than54%
of all ranked models
Faster than17%
of all ranked models
Cheaper than58%
of all ranked models

Overview

DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's...

Benchmarks

independent · via OpenRouter
Artificial Analysis51th percentile
Intelligence Index
28.5
Coding Index
31.9
Agentic Index
28.6
GPQA Diamond
75%
Humanity's Last Exam
8%
SciCode
32%
Tau²-Bench (agentic)
37%
Design Arena · Elo7,784 tournaments
UI Component
1238
Website
1232
codecategories
1227
3D
1221
Data Viz
1208
Game Dev
1200
svg
1122

Providers & pricing (5)

ProviderIn $/MOut $/MUptime
DeepInfrafp4$0.270$0.95096.1%
SiliconFlowfp8$0.270$1.0098.8%
Novitafp8$0.270$1.00100%
AtlasCloudfp8$0.300$0.950100%
StreamLake$0.343$1.03

Specifications

Context window164K
Max output33K
Knowledge cutoffMar 2025
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price$0.130/M
ModeratedNo

DeepSeek V3.1 Terminus FAQ

How much does DeepSeek V3.1 Terminus cost?

DeepSeek V3.1 Terminus costs $0.270 per million input tokens and $0.950 per million output tokens via OpenRouter, making it 124th cheapest of 298 paid models.

How smart is DeepSeek V3.1 Terminus?

DeepSeek V3.1 Terminus scores 28.5 on the Artificial Analysis Intelligence Index, ranking 82nd of 180 benchmarked models, with a GPQA Diamond score of 75%.

How fast is DeepSeek V3.1 Terminus?

DeepSeek V3.1 Terminus generates around 26 tokens per second with 1.1s time-to-first-token (p50), the 254th fastest tracked model.

What is DeepSeek V3.1 Terminus's context window?

DeepSeek V3.1 Terminus supports a 164K-token context window and can output up to 33K tokens. It accepts text input.

Compare head-to-head