DeepSeek V3.1 Terminus

Newer version available: DeepSeek V4 Pro

Released
2025-09-22
Last refreshed
2026-10-17
Status
Researched 28d ago
Open weightsCommercial use: permittedRAGLong contextClassificationJSON / Tool use
DeepSeek releases · 10 in the last 12 months · this family litChangelog →
Specifications
Family
DeepSeek
Released
2025-09-22
Context
163.8K
Parameters
671B total, 37B active (MoE)
Architecture
Decoder Only
Specialization
general
Openness
Open weights
License
DeepSeek LicensePermitted
Weights
Unknown
Code
Unknown
Created by
DeepSeek

Advancing artificial general intelligence (AGI).

Hangzhou, Zhejiang, China
Founded 2023
Website
Pricing
Output / 1M
$0.790
Input / 1M
$0.210

Cheapest of 3 routes · OpenRouter

Providers(3)
NVIDIA NIMOpenRouterNovita AI
View 3 provider routes
Links
Website

About

DeepSeek: DeepSeek V3.1 Terminus available via OpenRouter. Pricing: $0.21/1M input, $0.79/1M output.

Providers(3)

Compare all 3

Compare API pricing across 3 tracked provider routes for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
OpenRouter$0.210$0.790
Serverless
Novita AI$0.270$1.00
Serverless
NVIDIA NIM--
ServerlessPartial

Deprecated provider routes(1)

ProviderStatusCatalog ID
Vercel AI Gateway
DeprecatedRetired November 2, 2026
deepseek/deepseek-v3.1-terminus

Available via routers & gateways(1)

NVIDIA LLM Router Blueprint

Router

NVIDIA's open-source AI blueprint for LLM routing that selects the optimal model per prompt via intent classification or neural auto-routing; deprecated in favor of NVIDIA NeMo Switchyard.

Free OSSNVIDIA NIM

Capabilities

Structured Outputs

Benchmark peer barsfor RAG

No task-mapped benchmark peers are available for this model yet.

Benchmark scores(2)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
Chatbot Arena
1416.0text_style_control overall 2026-10-22 (style control, default view)Observed 2026-10-26
Configuration: DeepSeek-V3.1-Terminus (Thinking)
Harness: LMArena blind pairwise human votes, Bradley-Terry
Evaluator: LMArena
Confidence: confirmed
Notes: LMArena text style control leaderboard, overall category, published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split). Board entry 'DeepSeek-V3.1-Terminus (Thinking)': rank 139, Arena Score 1416.2 (95% CI 1406.1-1426.3), 3375 votes. Retrieved 2026-10-26.
Source

Source: LMArena (lmarena.ai), CC BY 4.0

LMArena Text Arena - Creative Writing
1391.0text creative_writing 2026-10-22Observed 2026-10-22
Configuration: DeepSeek-V3.1-Terminus (Thinking)
Harness: LMArena blind pairwise human votes, Bradley-Terry
Evaluator: LMArena
Confidence: confirmed
Notes: LMArena Text arena, Creative Writing category (no style control), board published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @a467e6b9, text config, latest split). Board entry 'DeepSeek-V3.1-Terminus (Thinking)' (category creative_writing in text/latest-00000-of-00001.parquet): rank 115 of 416, rating 1390.8 (95% CI 1363.1-1418.5), 435 votes. Other board entries for this model, not used: 'DeepSeek-V3.1-Terminus' 1403.4 (rank 91). Source: LMArena (CC BY 4.0). Retrieved 2026-10-27.
Source

Source: LMArena (lmarena.ai), CC BY 4.0

Migration checks

No linked migration route is available for this model yet.