GPT-5.2 vs GPT-6 Astra
| Spec | GPT-5.2 | GPT-6 Astra |
|---|---|---|
| Respan Index | 46.7 (#51) | 72.3 (#7) |
| Blended price per 1M | $4.81 | $20 |
| Context window | 400K | 1.1M |
| Input / output per 1M | $1.75 / $14 | $10 / $50 |
| Max output | 128K | 128K |
| Input | text, vision | text, vision |
| Released | 2025-12-11 | 2026-09-03 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | GPT-5.2 | GPT-6 Astra |
|---|---|---|
| Coding | 26.0 GPT-5.2 | 58.5 GPT-6 Astra |
| Agents | 59.5 GPT-5.2 | 78.3 GPT-6 Astra |
| Reasoning | 55.3 GPT-5.2 | 86.0 GPT-6 Astra |
| Knowledge & language | 38.1 GPT-5.2 | 75.6 GPT-6 Astra |
| Human preference | 51.5 GPT-5.2 | 60.1 GPT-6 Astra |
Benchmarks both report
GPT-6 Astra leads on 24 of the 25 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | GPT-5.2 | GPT-6 Astra |
|---|---|---|
| LMArena WebDev (LMArena) | 1419 | 1764 |
| LMArena Elo | 1437 | 1475 |
| LMArena Vision (LMArena) | 1243 | 1281 |
| MirrorCode (Epoch AI) | 32.2% | 60% |
| Chess Puzzles (Epoch AI) | 48% | 76% |
| EBR-bench (Epoch AI) | 23% | 76.2% |
| FrontierMath Tier 4 v2 (Epoch AI) | 31.7% | 97.6% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 67.4% | 93.7% |
| Furniture Assembly (Epoch AI) | 38.3% | 80% |
| GPQA Diamond | 92.4% | 96% |
| LiveBench Agentic Coding (LiveBench) | 50.3% | 57.3% |
| LiveBench Coding (LiveBench) | 76.1% | 80.4% |
| LiveBench Data Analysis (LiveBench) | 78.2% | 83% |
| LiveBench Instruction Following (LiveBench) | 61.8% | 75.6% |
| LiveBench Language (LiveBench) | 79.8% | 89.4% |
| LiveBench Mathematics (LiveBench) | 93.2% | 96.8% |
| LiveBench Reasoning (LiveBench) | 83.2% | 92.7% |
| Mystery Game Puzzles (Epoch AI) | 23% | 84% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 96.1% | 100% |
| SAGE (Vals AI) | 49.3% | 46.4% |
| SimpleQA Verified | 37.1% | 75.6% |
| Vending-Bench 2 (Andon Labs) | 3591.33 | 15514.7 |
| WeirdML (Håvard Tveit Ihle) | 72.2% | 93.6% |
| GSO Opt@1 (GSO) | 26.5% | 77.5% |
| SimpleBench (SimpleBench) | 45.8% | 86.5% |
FAQ
Is GPT-5.2 better than GPT-6 Astra?
GPT-6 Astra scores 72.3 on the Respan Index (#7), ahead of GPT-5.2 at 46.7 (#51). GPT-6 Astra leads on 24 of the 25 benchmarks both report.
Which is cheaper, GPT-5.2 or GPT-6 Astra?
GPT-5.2 lists at $1.75 input and $14 output per 1M tokens; GPT-6 Astra at $10 and $50. GPT-5.2 is 4.2x cheaper per token at list price.
Which has the larger context window?
GPT-5.2 accepts 400K tokens and GPT-6 Astra 1.1M tokens.