Are the weights open and can I run this locally?

Yes — DeepSeek-V3.1-Terminus is released under the MIT licence and the weights are on Hugging Face at deepseek-ai/DeepSeek-V3.1-Terminus. The model structure is the same as DeepSeek-V3, so the same local inference tooling applies, though the model is large and requires substantial hardware.

What does it cost and what is the context window?

Via API the price is $0.27 per million input tokens and $1.00 per million output tokens, with a 163,840-token context window.

How is Terminus different from DeepSeek V3.1?

Terminus is a patch release, not a new model — it targets two specific weaknesses: language consistency (less mixed Chinese-English output) and agent performance (Code Agent and Search Agent). Most benchmark scores are within a point of V3.1, but agentic tasks like SWE-bench and Terminal-bench show the clearest gains.

DeepSeek V3.1 Terminus

1961°

#llm #deepseek #reasoning #agents #coding #local

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

DeepSeek | 163,840 token context | $0.27/M input, $1.00/M output. DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's...

Guides for this model

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

Open weights vs open source vs API-only: what you are actually allowed to do

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

DeepSeek-V3.1-Terminus is a targeted update to DeepSeek V3.1 from DeepSeek, fixing two specific complaints: mixed-language output and weak agent performance. Benchmark gains are modest on most tasks but meaningful on agentic ones — SWE-bench, BrowseComp and Terminal-bench all move up. It suits developers building code or search agents who were already using V3.1 and hit those rough edges.

Known for

Stronger Code Agent and Search Agent performance over V3.1
Reduced mixed Chinese-English output and stray characters
Solid SWE-bench scores: 68.4% verified, 57.8% multilingual
MIT licence — weights are open and free to use commercially
163,840-token context window at well under $1.50/M combined

Watch out for

A known FP8 scale data format issue in self_attn.o_proj exists; DeepSeek says a fix is coming.
This is a large model — running it locally demands serious hardware; check VRAM requirements before downloading.
Codeforces rating dips slightly versus V3.1, so pure competitive-coding users may not see an improvement.

Best for

Prompts to try

Terminal task agent

You have access to a terminal. A Python project in /app is failing its test suite. Run the tests, read the output, identify the root cause, patch the relevant file, and run the tests again to confirm the fix. Show each command you run and explain your reasoning between steps.

Multi-file refactor

I have a Flask app where database calls are scattered across route handlers. Refactor it so all DB access goes through a single repository layer. Show the new file structure, the repository module, and two example route handlers rewritten to use it. Keep the existing behaviour.

Hard reasoning puzzle

Five engineers each own a different programming language, drink a different beverage, and sit in a row of five desks. The Rust engineer sits next to the tea drinker. The Python engineer is at one end. The coffee drinker is two seats from the Go engineer. The TypeScript engineer drinks water. Who drinks espresso, and what language do they use? Show your reasoning.

Compare with

DeepSeek V4.1 Flash
Open-weights model
1,048,576 token context · $0.04/M input, $0.68/M output
DeepSeek V4 Flash 0731
Open-weights model
1,048,576 token context · $0.01/M input, $1.28/M output
DeepSeek V4 Flash 0423
Open-weights model
1,048,576 token context · $0.09/M input, $0.18/M output
DeepSeek V4 Pro 0813
Open-weights model
1,048,576 token context · $1.32/M input, $3.96/M output

More text models from deepseek

Questions people ask

Are the weights open and can I run this locally?: Yes — DeepSeek-V3.1-Terminus is released under the MIT licence and the weights are on Hugging Face at deepseek-ai/DeepSeek-V3.1-Terminus. The model structure is the same as DeepSeek-V3, so the same local inference tooling applies, though the model is large and requires substantial hardware.
What does it cost and what is the context window?: Via API the price is $0.27 per million input tokens and $1.00 per million output tokens, with a 163,840-token context window.
How is Terminus different from DeepSeek V3.1?: Terminus is a patch release, not a new model — it targets two specific weaknesses: language consistency (less mixed Chinese-English output) and agent performance (Code Agent and Search Agent). Most benchmark scores are within a point of V3.1, but agentic tasks like SWE-bench and Terminal-bench show the clearest gains.

Captain's verdict
A sensible patch for V3.1 users who hit language-mixing bugs or needed better agent reliability — not a reason to switch if you did not.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

DeepSeek V3.1 Terminus is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm deepseek reasoning agents coding local

Browse by Model Family