Up to four models
Qwen3.8-27B vs GPT-5.2-Codex
On the index (GPT-4 = 100), Qwen3.8-27B scores 155 and GPT-5.2-Codex 152. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
155139–171
WeightsOpen
ReleasedAug 2026
Price in / out$0.060 / $2.2
Context1.02M
Core tests taken3 of 9
GPT-5.2-Codex
152135–169
WeightsClosed
ReleasedDec 2025
Price in / out$1.75 / $14
Context400K
Core tests taken0 of 9
Qwen3.8-27B and GPT-5.2-Codex are too close to call (3 points apart, within the uncertainty), and costs 8.1× less.
Test by testthird-party results
Every result as a table
| Test | Qwen3.8-27B | GPT-5.2-Codex |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 76.6% | 78.2% |
| LiveBench IF | 72.7% | 66.5% |
| LiveBench Language | 74.4% | 73.7% |
| LiveBench Reasoning | 80.0% | 77.7% |
| Maths | ||
| LiveBench Mathematics | 86.2% | 88.8% |
| Coding | ||
| LiveBench Agentic Coding | 61.4% | 49.4% |
| LiveBench Coding | 75.7% | 83.6% |
| LiveCodeBench | 84.0% | 88.0% |
| SWE-bench | 86.0% | 72.4% |
| Vibe Code Bench v1.1 | 64.8% | 37.9% |
Tests at least two of these models have taken. Bold rows are the core tests.