DeepSeek: DeepSeek V3.1
Analysis Summary
DeepSeek V3.1 is the latest iteration of DeepSeek's V3 series, offering tool use and function calling across a 163K context window. With a coding index of 29.7, math index of 89.7, and GPQA of 0.779, it shows strong analytical and coding capability for its price tier. The intelligence index of 27.7 places it in the upper-mid range, and the livecodebench score of 0.784 is competitive with models costing significantly more.
For businesses, V3.1 is well-suited for code generation, mathematical analysis, and structured document workflows. The 163K context window supports long-document processing, and the pricing at $0.21/1M input and $0.79/1M output offers strong value for the capability delivered. The agentic index of 31.2 is adequate for straightforward tool-use pipelines but not for complex autonomous agents.
A -4 point regional accessibility adjustment applies. Teams comfortable with DeepSeek's deployment model will find V3.1 a cost-effective coding and analytical workhorse; those requiring enterprise-grade support from major providers should factor that into their evaluation.
Assessed June 6, 2026
Editorial notes
DeepSeek V3.1 delivers strong coding and math benchmarks with a 163K context window and tool use at $0.21/1M input, though provider accessibility may limit enterprise adoption for some teams.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How DeepSeek: DeepSeek V3.1 compares
DeepSeek: DeepSeek V3.1 ranks #126 of 377 AI models we track for overall intelligence, #90 of 314 for coding, #137 of 289 for agentic tasks. Its 164K-token context window is larger than 62% of the models we list. At $0.21 per million input tokens it is cheaper than 49% of comparable models.
About DeepSeek: DeepSeek V3.1
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context..
Capabilities
Architecture Detail
| Instruct Type | deepseek-v3.1 |
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does DeepSeek: DeepSeek V3.1 stack up?
Compare side-by-side with other professional models.
Model Information
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.21 | $0.000210 |
| Output | $0.79 | $0.000790 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about DeepSeek: DeepSeek V3.1
How much does DeepSeek: DeepSeek V3.1 cost?
DeepSeek: DeepSeek V3.1 costs $0.21 per million input tokens and $0.79 per million output tokens.
What is the context window of DeepSeek: DeepSeek V3.1?
DeepSeek: DeepSeek V3.1 has a context window of 163,840 tokens (164K).
Is DeepSeek: DeepSeek V3.1 good for coding?
On our coding benchmark index, DeepSeek: DeepSeek V3.1 ranks #90 of 314 models, placing it in the broader range of the field for code generation and debugging.
What can DeepSeek: DeepSeek V3.1 do?
DeepSeek: DeepSeek V3.1 supports tool use and function calling.
Who created DeepSeek: DeepSeek V3.1?
DeepSeek: DeepSeek V3.1 is developed by DeepSeek and was released on August 21, 2025.
