DeepSeek V3.1 Terminus
DeepSeek-V3.1-Terminus. The latest update builds on V3.1's strengths while addressing key user feedback. Language consistency improvements (fewer CN/EN mix-ups, no random chars), stronger Code Agent & Search Agent performance, and more stable, reliable outputs across benchmarks. FP8.
- Tool Calling
- Structured Output
Added Aug 2, 2025
Model weightsPricing
Auto routing · per 1M tokens- Input
- $0.25
- Output
- $0.70
- Cache read
- $0.125
Specifications
- Context window
- 128K
- Max output
- 65.5K
- Parameters
- 671B / 37B
- Total / active
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1414.7
Overall Rank
#145 / 421
Votes
3,605
Confidence Interval
1405.0 - 1424.4
Category Scores
Coding
#173 / 416
769 votes
1440.2
Math
#170 / 404
216 votes
1396.4
Longer Query
#147 / 399
838 votes
1419.0
Creative Writing
#105 / 419
446 votes
1404.8
Instruction Following
#164 / 421
1,016 votes
1394.1
Hard Prompts
#160 / 421
1,898 votes
1424.1
Additional Categories14
Industry Medicine And Healthcare
#100 / 390
259 votes
1456.6
Industry Life And Physical And Social Science
#112 / 419
570 votes
1451.6
Industry Writing And Literature And Language
#115 / 420
789 votes
1406.8
Russian
#121 / 385
304 votes
1420.9
Non English
#135 / 421
2,025 votes
1404.1
Polish
#144 / 244
229 votes
1397.2
Industry Entertainment And Sports And Media
#148 / 419
625 votes
1377.6
Industry Legal And Government
#149 / 393
263 votes
1417.7
Exclude Ties
#151 / 421
2,581 votes
1400.1
English
#159 / 421
1,580 votes
1417.7
Hard Prompts English
#173 / 419
848 votes
1424.4
Industry Business And Management And Financial Operations
#173 / 414
717 votes
1396.6
Industry Software And It Services
#175 / 421
1,277 votes
1428.5
Multi Turn
#178 / 419
658 votes
1394.2
Published 2026-11-09 · Matched as DeepSeek-V3.1-Terminus · CC BY 4.0
LMArena DatasetLeaderboard results
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026 | ||
UGI Score | 57.6 | #14 of 149 |
UGI Willingness | 6.0 | #26 of 149 |
UGI Writing | 54.0 | #51 of 149 |
Filtered and ranked to models on NanoGPT.
Providers
Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)
 | $0.25 | $0.70 | $0.125 | N/A | N/A |
| $0.284 | $1.05 | N/A | N/A | N/A |
Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.