DeepSeek V3.1 Terminus
DeepSeek V3.1 Terminus by DeepSeek is a refinement of V3.1 fixing language mixing and strengthening Code Agent and Search Agent performance. MIT licensed.
Model Overview
Capabilities, design details, and architectural traits
DeepSeek V3.1 Terminus - Stability-Focused Refinement of DeepSeek V3.1
DeepSeek V3.1 Terminus is an update to DeepSeek V3.1, released as a targeted refinement addressing two officially documented problem areas from V3.1's initial deployment: language consistency and agent capability. It maintains the same architecture and hybrid thinking/non-thinking design as V3.1, with no structural changes.
Documented Improvements Over V3.1
Two officially documented fix areas:
Language consistency:
- Reduced occurrences of Chinese-English mixing in outputs
- Reduced occurrences of occasional abnormal characters
Agent capabilities:
- Further optimized performance of the Code Agent
- Further optimized performance of the Search Agent.
API Aliases
Upon release, both deepseek-chat and deepseek-reasoner API aliases were upgraded to DeepSeek V3.1-Terminus: deepseek-chat → non-thinking mode; deepseek-reasoner → thinking mode.
Benchmark Performance
Independent evaluations · Artificial Analysis
Accuracy & Capability Details
Compare Models Side-by-Side
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | DeepSeek | Anthropic | Anthropic |
| Release Date | September 22, 2025 | September 22, 2026 | September 28, 2026 |
| Knowledge Cutoff | Mar 2025 | - | Jun 2026 |
| Context & Limits | |||
| Context Window | 164K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $0.27 Best Input Pricing | $4 | $2 |
| Output Pricing | $1 Best Output Pricing | $20 | $10 |
| Modalities | |||
| Inputs | text | textimagefile | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 13.9 | 57.6 Best Intelligence Index | 56.0 |
| Coding Index | - | - | - |
| Agentic Index | - | - | - |
Humanity's Last Exam
Extremely difficult logical reasoning and knowledge.
Long Context Reasoning
Logical reasoning over long context windows.
SciCode Benchmark
Scientific coding and mathematical modeling.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Explore more from DeepSeek
Other models by DeepSeek
DeepSeek V4.1 Flash
DeepSeek V4 Flash Vision
DeepSeek V4 Pro
DeepSeek V4 Flash
Top AI Models
Leading alternatives by intelligence score
