---
title: "DeepSeek Coder 2 236B Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for DeepSeek Coder 2."
source: "https://rankedagi.com/models/deepseek-coder-2-236b"
---

# DeepSeek Coder 2

| Field | Value |
| --- | --- |
| Organization | DeepSeek |
| License | Proprietary |
| Size | 236B |
| Version | Latest |
| Released | 2024-07-22 |
| Context window | 128K |
| Knowledge cutoff | 2023-11-01 |
| Input cost per million tokens | $0.14 |
| Output cost per million tokens | $0.28 |
| Cached input cost per million tokens | $0.01 |
| Last updated | 2026-05-03T23:59:41.187Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 27.7% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 21.9% | RankedAGI Agentic Score |  |
| ChatArena (LMSYS) | coding | 1226 | ChatArena (LMSYS) Coding ELO Score with Style Control |  |
| LiveBench Coding Score (old) | coding | 41.5% | LiveBench Coding Score old version |  |
| RankedAGI Reasoning | reasoning | 32.8% | RankedAGI Reasoning Score |  |
| GPQA Diamond | reasoning | 42.0% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1264 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| HumanEval | coding | 90.2% | HumanEval coding benchmark |  |
| HumanEval+ | coding | 82.3% | HumanEval+ benchmark |  |
| MMLU Pro | knowledge | 63.6% | Graduate-level visual problem-solving |  |
| MMLU | knowledge | 80.0% | Image Reasoning |  |
| LiveBench Average | general | 47.2% | LiveBench Average Score old version |  |
| LiveBench Data Analysis | general | 38.3% | LiveBench Data Analysis Score | https://livebench.ai/ |
| LiveBench Language | general | 33.0% | LiveBench Language Score | https://livebench.ai/ |
| Artificial Analysis Quality Index | general | 67 | Artificial Analysis Quality Index (third-party composite, deprecated) |  |
| RankedAGI Math | math | 39.4% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 35.1% | Overall RankedAGI score |  |