BenchLeader

Deepseek v2.5

Quality#215 of 339
45.8Index
Below the median
−4.4 on the median

As of 16 Nov 2026, Deepseek v2.5 by DeepSeek scores 45.8 on the BenchLeader Index (608 ranked configurations), with its best results in reasoning (46) and weakest in coding (36). Released 6 Sep 2024.

Key facts (as of 16 Nov 2026)

  • Deepseek v2.5 was released on 6 Sep 2024 (DeepSeek).

In detail. Deepseek v2.5 is a DeepSeek open-weights model, released 6 Sep 2024. Its best configuration ranks #444 of 608 on the BenchLeader Index at 45.8 ±5.7, in the lower half. The ± is the point: 251 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in reasoning (46) and lowest in coding (36). Last measured 11 Nov 2026.

Blended price
–
Output speed
–
First answer
–
Context
128k
via Artificial Analysis
Cost per task
–
Tokens per task
–
Released
6 Sep 2024
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index46
  2. Reasoning46
  3. Coding36
  4. Maths45
  5. Human preference46

Embed this rank

More badge styles
[![Deepseek v2.5 on BenchLeader: #211 overall](https://www.benchleader.com/badge/deepseek-v2-5.svg?style=compact)](https://www.benchleader.com/models/deepseek-v2-5)

Paste it at the top of README.md, beside your other badges. The badge redraws with every data update and links back to this page.

Get an email when DeepSeek ships a new model

And once when it gets ranked, plus the weekly review on Mondays.

Choose providers and emails

Versions

DeepSeek has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

ModelReleasedIndexRank
Deepseek v2.5this page6 Sep 202445.8#444
Deepseek v27 May 2024––

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmarknot statedSourceTrend
Arena Hard Prompts1321#251Arena (LMArena)
GPQA Diamond (AA)not in index42.3%#477Artificial Analysis
Arena Expertnot in index1296#255Arena (LMArena)

Coding

Benchmarknot statedSourceTrend
Arena Coding1368#235Arena (LMArena)
Aider Polyglot17.8%#53Aider polyglot leaderboard

Maths

Benchmarknot statedSourceTrend
Arena Maths1287#246Arena (LMArena)

Instruction following

Benchmarknot statedSourceTrend
Arena Instruction Followingnot in index1291#257Arena (LMArena)

Human preference

Benchmarknot statedSourceTrend
Arena Text1307#258Arena (LMArena)
Arena Creative Writingnot in index1265#262Arena (LMArena)
Arena Multi-turnnot in index1292#258Arena (LMArena)
Arena Longer Queriesnot in index1314#256Arena (LMArena)

Composite

Benchmarknot statedSourceTrend
AA Progress Index v1not in index160.1#531Artificial Analysis

More from DeepSeek

New this month

See also

Embed or cite

Show where Deepseek v2.5 ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page.

[![Deepseek v2.5 on BenchLeader: #211 overall](https://www.benchleader.com/badge/deepseek-v2-5.svg)](https://www.benchleader.com/models/deepseek-v2-5)

Cite this

The ranking changes daily, so a reference carries the data date. APA style:

BenchLeader. (2026, November 16). Deepseek v2.5 on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 16, 2026, from https://www.benchleader.com/models/deepseek-v2-5

BibTeX:

@misc{benchleader_deepseek_v2_5_20261116,
  author       = {{BenchLeader}},
  title        = {{Deepseek v2.5 on the BenchLeader Index (method v1.4)}},
  year         = {2026},
  month        = nov,
  howpublished = {\url{https://www.benchleader.com/models/deepseek-v2-5}},
  note         = {Data as of 2026-11-16}
}

Every badge style, the embeddable top-10 widget and how to cite the Index

Questions people ask

What are Deepseek v2.5's benchmark scores?
As of 16 Nov 2026, Deepseek v2.5 scores 45.8 ±5.7 on the BenchLeader Index, strongest in reasoning (46) and weakest in coding (36). That rests on independent benchmark results only, and every score on this page links to its publisher.
How does Deepseek v2.5 rank?
Deepseek v2.5 is #444 of 608 ranked configurations on 16 Nov 2026. 251 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Fable 5.2 (max effort) at 72.7.
When was Deepseek v2.5 released?
DeepSeek released Deepseek v2.5 on 6 Sep 2024.

Data as of 16 Nov 2026. Compare with another model.

Cite as: BenchLeader, “Deepseek v2.5: benchmarks, pricing, speed and rank”, data as of 16 Nov 2026, https://www.benchleader.com/models/deepseek-v2-5.