computetrail

All models ·deepseek

DeepSeek: DeepSeek V3.1 Terminus

Provider: deepseek · deepseek/deepseek-v3.1-terminus

Input
$0.27 /1M
Output
$1 /1M
Cache read
$0.135 /1M
Context
163,840 tokens

How DeepSeek: DeepSeek V3.1 Terminus pricing compares

At $1 / 1M output tokens, DeepSeek: DeepSeek V3.1 Terminus is the 7th-cheapest of 14 deepseek models we track. It's cheaper than 67% of the 392 paid models on computetrail (129th-cheapest overall). Output costs 3.7× its input price ($0.27 / 1M in). For a typical 3:1 input-to-output workload that works out to about $0.45 / 1M blended. Reading cached input costs 50% less than fresh input, so prompt caching cuts repeat-context cost sharply. Its 163,840 tokens context window ranks 270th-largest of 392 models with a published limit.

Output price history

$1.00$0.95006-2907-2908-27

Across 60 daily snapshots since 2026-06-29, the output price has risen 5.3% — from $0.95 to $1 / 1M.

Similar-priced models

ModelProviderOutput $/1MContext
Qwen: Qwen3 Coder 480B A35Bqwen$1262,144
Perplexity: Sonarperplexity$1127,072
MiniMax: MiniMax M2minimax$1.02204,800
DeepSeek: DeepSeek V3deepseek$1.0287163,840
Google: Gemini 3.7 Flash (batch)google$0.93751,048,576
MiniMax: MiniMax M2.5minimax$1.08204,800

Prices are recorded daily from public provider data via OpenRouter and normalized to USD per 1M tokens. See methodology.