█

AI/TLDR

Buy me a coffee
Buy me a coffee
PROMOTE YOUR WEBSITE IN JUST $20

Model comparison

DeepSeek-V3.1-Terminus vs DeepSeek-V3.1

DeepSeekDeepSeek-V3.1-Terminus
Input
$0.27 / 1M tokens
Output
$0.95 / 1M tokens
Context
128K
Released
2025-09-22
Full model page →DeepSeekDeepSeek-V3.1
Input
$0.21 / 1M tokens
Output
$0.79 / 1M tokens
Context
128K
Released
2025-08-21
Full model page →
Short answer

DeepSeek-V3.1-Terminus scores higher on 8 of the 9 benchmarks both models report (the other leads on 1); DeepSeek-V3.1 is cheaper on input tokens ($0.2 vs $0.3 per 1M, 1.3×).

Specs and pricing side by side

SpecDeepSeek-V3.1-TerminusDeepSeek-V3.1
MakerDeepSeekDeepSeeksame
Released2025-09-222025-08-21
Context window128K128Ksame
Max output32Knot published
Input price$0.27 / 1M tokens$0.21 / 1M tokens
Output price$0.95 / 1M tokens$0.79 / 1M tokens
Parameters685B total, 37B active (MoE)671B total / 37B active (MoE)
LicenseMITMITsame
WeightsOpen weightsOpen weightssame
ModalitiesTextTextsame

Not published by either maker: knowledge cutoff.

Benchmarks both models report

BenchmarkDeepSeek-V3.1-TerminusDeepSeek-V3.1
MMLU-Pro85%84.8%
GPQA-Diamond80.7%80.1%
Humanity's Last Exam21.7%15.9%
LiveCodeBench74.9%74.8%
Aider-Polyglot76.1%76.3%
SWE-bench Verified68.4%66%
SWE-bench Multilingual57.8%54.5%
Terminal-bench36.7%31.3%
BrowseComp38.5%30%

Scores are as published by each maker or its cited source; harness, reasoning effort and date can differ between rows, so treat small gaps as ties. Sources are listed on each model page.

Which one should you pick?

Pick DeepSeek-V3.1-Terminus if…

  • You care about MMLU-Pro, GPQA-Diamond, Humanity's Last Exam (it scores higher on 8 shared benchmarks).
  • Your use case is: agentic coding assistants and autonomous software-engineering loops (SWE-bench-style tasks).

Pick DeepSeek-V3.1 if…

  • You care about Aider-Polyglot (it scores higher on 1 shared benchmark).
  • Cost matters: input tokens cost 1.3× less ($0.2 vs $0.3 per 1M).
  • Your use case is: coding agents and software-engineering tasks (SWE-bench-style fix/patch workflows).

FAQ

Which is better, DeepSeek-V3.1-Terminus or DeepSeek-V3.1?

DeepSeek-V3.1-Terminus scores higher on 8 of the 9 benchmarks both models report (the other leads on 1); DeepSeek-V3.1 is cheaper on input tokens ($0.2 vs $0.3 per 1M, 1.3×). The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.

How much cheaper is DeepSeek-V3.1?

DeepSeek-V3.1-Terminus: $0.27 / 1M tokens input, $0.95 / 1M tokens output. DeepSeek-V3.1: $0.21 / 1M tokens input, $0.79 / 1M tokens output (list API prices as published by each maker).

What are the context windows of DeepSeek-V3.1-Terminus and DeepSeek-V3.1?

DeepSeek-V3.1-Terminus offers 128K with up to 32K output; DeepSeek-V3.1 offers 128K.

Is DeepSeek-V3.1-Terminus newer than DeepSeek-V3.1?

DeepSeek-V3.1-Terminus was released 2025-09-22; DeepSeek-V3.1 was released 2025-08-21.

Go deeper