---
title: "DeepSeek V3.1 Think 840B Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for DeepSeek V3.1 Think."
source: "https://rankedagi.com/models/deepseek-v3-1-reasoner"
---

# DeepSeek V3.1 Think

| Field | Value |
| --- | --- |
| Organization | DeepSeek |
| License | Open Source |
| Type | Thinking |
| Size | 840B |
| Version | Latest |
| Released | 2025-08-21 |
| Context window | 128K |
| Input cost per million tokens | $0.56 |
| Output cost per million tokens | $1.68 |
| Cached input cost per million tokens | $0.07 |
| Model ID | deepseek-reasoner |
| Last updated | 2026-05-03T23:59:41.186Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 60.0% | RankedAGI Coding Score |  |
| SWEBench Multilingual | coding | 54.5% |  |  |
| SWEBench Verified | coding | 66.0% | Agentic Coding | https://www.swebench.com/index.html |
| RankedAGI Agentic | Agents | 48.6% | RankedAGI Agentic Score |  |
| BrowseComp | Agents | 30.0% | A benchmark for browsing agents | https://openai.com/index/browsecomp/ |
| LiveCodeBench v5 | coding | 73.3% | LiveCodeBench - Evaluation of Large Language Models for Code |  |
| Codeforces ELO Score | coding | 2091 | Codeforces ELO Score |  |
| Aider Polyglot | coding | 76.3% | Aider Polyglot Code Completion Benchmark |  |
| RankedAGI Reasoning | reasoning | 63.3% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 29.8% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 81.0% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1417 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| AIME 2025 | math | 87.5% | AIME 2025 Competition Math |  |
| AIME 2024 | math | 93.1% | AIME 2024 Competition Math |  |
| MMLU Pro | knowledge | 84.8% | Graduate-level visual problem-solving |  |
| MMLU | knowledge | 93.7% | Image Reasoning |  |
| RankedAGI Math | math | 73.5% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 56.4% | Overall RankedAGI score |  |
| GDPval-AA Elo | Agents | 1007 | Office Tasks (Artificial Analysis) |  |
| Code DesignArena | Design | 1240 |  | https://www.designarena.ai/leaderboard/code |