ModelsDeepSeek V3.1

DeepSeek V3.1

by DeepSeek

Cost-effective

Best value for production workloads

Pricing
Input
$0.56
per 1M tokens
Output
$1.68
per 1M tokens
Cached
$0.07
per 1M tokens
Note: Released Aug 2025. MIT license. Beats GPT-4 at 17% of cost
Context & Output
Context Window128K tokens
Max Output32K tokens
Latency
Fast
Capabilities
Multimodal
Streaming
Function Calling
Prompt Caching
Key Strengths
What makes this model stand out
Hybrid thinking/non-thinking
671B params
82.6% HumanEval
Try DeepSeek V3.1Compare All Models
Similar Models in Cost-effective Tier
Other models with similar pricing and performance characteristics
GPT-5 Mini
OpenAI
Input:$0.25/M
Context:272K tokens
View Details
DeepSeek R1
DeepSeek
Input:$0.55/M
Context:128K tokens
View Details
Qwen3-235B
Alibaba
Input:$0.35/M
Context:128K tokens
View Details