Command Palette

Search for a command to run...

DeepSeek V3.1 Terminus (Non-reasoning)

DeepSeek

MIT

Analysis of DeepSeek's DeepSeek V3.1 Terminus (Non-reasoning) and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more.

Context

Release Date
Sep 22, 2025
Window
128k

PricingPer 1M tokens

Input
$0.27
Output
$1
Blended 3:1
$0.4525

Capabilities

Input
Output
Reasoning tokens

Benchmarks

Intelligence
●○○○○
Math
●●●○○
Coding
●●○○○
MMLU Pro
83.6%
GPQA
75.1%
HLE
8.4%
SciCode
32.1%
AIME
53.7%
LiveCodeBench
52.9%

Analysis of DeepSeek's DeepSeek V3.1 Terminus (Non-reasoning) and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more.