---
title: "Deepseek 2.5 Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Deepseek 2.5."
source: "https://rankedagi.com/models/deepseek-2.5-old"
---

# Deepseek 2.5

| Field | Value |
| --- | --- |
| Organization | DeepSeek |
| License | Proprietary |
| Version | Old |
| Released | 2024-12-10 |
| Context window | 128K |
| Knowledge cutoff | 2023-11-01 |
| Last updated | 2026-05-03T23:59:41.187Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 40.0% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 46.5% | RankedAGI Agentic Score |  |
| LiveCodeBench v6 | coding | 34.4% | LiveCodeBench v6 - Evaluation of Large Language Models for Code |  |
| ChatArena (LMSYS) | coding | 1262 | ChatArena (LMSYS) Coding ELO Score with Style Control |  |
| RankedAGI Reasoning | reasoning | 40.9% | RankedAGI Reasoning Score |  |
| Text Arena | general | 1323 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| Hughes Hallucination Evaluation | safety | 2.4% | Computed using Vectara's Hallucination Evaluation Model, also known as HHEM. This evaluates how often an LLM introduces hallucinations when summarizing a document. | https://github.com/vectara/hallucination-leaderboard |
| RankedAGI Math | math | 60.4% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 46.0% | Overall RankedAGI score |  |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://api-docs.deepseek.com/news/news1210 |