---
title: "Qwen 3.5 27B Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Qwen 3.5."
source: "https://rankedagi.com/models/qwen-3-5-27b"
---

# Qwen 3.5

| Field | Value |
| --- | --- |
| Organization | Alibaba |
| License | Open Source |
| Type | Thinking |
| Size | 27B |
| Version | Latest |
| Released | 2026-02-24 |
| Last updated | 2026-05-04T01:24:28.741Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 63.8% | RankedAGI Coding Score |  |
| SWEBench Pro | coding | 51.2% | Diverse Agentic Coding |  |
| SWEBench Multilingual | coding | 69.3% |  |  |
| SWEBench Verified | coding | 75.0% | Agentic Coding | https://www.swebench.com/index.html |
| Terminal-Bench 2.0 | coding | 41.6% | Agentic Terminal Coding | https://www.tbench.ai/leaderboard/terminal-bench/2.0 |
| RankedAGI Agentic | Agents | 53.1% | RankedAGI Agentic Score |  |
| Code Arena | coding | 1347 | Code Arena Elo Score | https://arena.ai/leaderboard/code |
| LiveCodeBench v6 | coding | 80.7% | LiveCodeBench v6 - Evaluation of Large Language Models for Code |  |
| RankedAGI Reasoning | reasoning | 62.4% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 24.3% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 85.5% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| AIME 2026 | math | 92.6% | AIME 2026 Competition Math | https://matharena.ai/?comp=aime--aime_2026&view=problem |
| Vending Bench 2 | Agents | $201.98 | Benchmark for measuring AI model performance on running a business over long time horizons. Models are tasked with running a simulated vending machine business over a year and scored on their bank account balance at the end. | https://andonlabs.com/evals/vending-bench-2 |
| MMLU Pro | knowledge | 86.1% | Graduate-level visual problem-solving |  |
| RankedAGI Math | math | 74.2% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 57.8% | Overall RankedAGI score |  |
| Toolathlon | Agents | 31.5% | Benchmark to assess language agents' general tool use in realistic environments. | https://toolathlon.xyz/introduction |

## Model Links

| Title | URL |
| --- | --- |
| HF | https://huggingface.co/Qwen/Qwen3.5-27B |