deepseek Open weights Reasoning

DeepSeek V3.1 Terminus pricing & benchmarks

Released Sep 22, 2025 · served by 5 providers · ranked #53 by intelligence in our index

Input / 1M tokens$0.270
Output / 1M tokens$0.950
Context window164K
Cached input$0.135
Intelligence index30.4
Coding index43.5
Agentic index18.1
1M in + 300K out$0.56

What DeepSeek V3.1 Terminus is

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...

Providers and prices for DeepSeek V3.1 Terminus

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
DeepInfracheapest $0.270 $0.950 164K fp4 99.5%
AtlasCloud $0.300 $0.950 131K fp8 95.6%
SiliconFlow $0.270 $1.00 164K fp8 89.2%
Novita $0.270 $1.00 131K fp8 100.0%
StreamLake $0.343 $1.03 128K 88.8%

DeepSeek V3.1 Terminus benchmark results

Design Arena head-to-head results, as reported through the OpenRouter model API.

CategoryArenaRankEloWin rate
uicomponent models #41 1218 59.3%
3d models #45 1198 56%
codecategories models #45 1206 56%
website models #46 1212 56.4%
dataviz models #52 1193 53%
svg models #52 1115 50.1%
gamedev models #58 1187 52.5%

Cheaper models in the same class

Models scoring within 4 points of DeepSeek V3.1 Terminus on the Intelligence Index, but with a lower output price.

FAQ

How much does the DeepSeek V3.1 Terminus API cost?

$0.270 per 1M input tokens and $0.950 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $0.56.

Is DeepSeek V3.1 Terminus free?

No. It is a paid model starting at $0.270 per 1M input tokens, though some providers offer trial credits.

Which provider is cheapest for DeepSeek V3.1 Terminus?

DeepInfra at $0.270 input / $0.950 output per 1M tokens (fp4 quantization).

Can I self-host DeepSeek V3.1 Terminus?

Yes — weights are published as deepseek-ai/DeepSeek-V3.1-Terminus on Hugging Face, so you can run it on your own hardware.