> DeepSeek-V2.5 (Sep 2024) is an open-weights AI model from DeepSeek, released 6 Sep 2024.

[Models](https://themodelindex.org/models/) /[DeepSeek](https://themodelindex.org/labs/deepseek/)

# DeepSeek-V2.5 (Sep 2024)

DeepSeek

Released 6 Sep 2024

Open

[Compare](https://themodelindex.org/compare/deepseek-v2-5-sep-2024-vs-o1-mini/)

–

Index score

Not yet evaluated independently

0/ 8

Core tests taken

0 of 6 areas · plus 1 anchor tests

–

Versus the frontier at release

Replaced by [DeepSeek-V3](https://themodelindex.org/models/deepseek-v3/)

–

Price per million tokens

no public API price found

–

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

Claude 3.5 Sonnet (Jun 2024) reached this score 3 months earlier.

## Test results · by area · tick = best by any model

Show 1 anchor tests

Reasoning & knowledge

no evidence yet

GPQA Diamond

no result

Humanity's Last Exam

no result

Maths

no evidence yet

FrontierMath T1-3

no result

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

no evidence yet

APEX-Agents

no result

Novel problems

no evidence yet

ARC-AGI-2

no result

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

LMArena · Coding

coding questions in chat

#224

of 383 · 1368

LMArena · Maths

maths questions in chat

#238

of 373 · 1287

LMArena · Text

overall chat quality, judged by people

#251

of 388 · 1307

LMArena · Creative writing

creative writing

#257

of 386 · 1265

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Replaced by DeepSeek-V3; it no longer leads any category among current models.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Facts and sources

Released

6 Sep 2024

Announcement

not linked yet

Release entry

–

Scored as

DeepSeek-V2.5 (Sep 2024)

Artificial Analysis index

–

Output speed

–

Hugging Face

not linked

## DeepSeek releases · around this one

DeepSeek Coder

Open

2 NOV 2023

DeepSeek LLM 67B

Open

29 NOV 2023

index 82

---
Source: https://themodelindex.org/models/deepseek-v2-5-sep-2024/ · The Model Index · data as of 2026-10-01