Claude Opus 4.7 vs GPT-5.4
| Spec | Claude Opus 4.7 | GPT-5.4 |
|---|---|---|
| Respan Index | 50.1 (#42) | 50.0 (#43) |
| Blended price per 1M | $10 | $5.63 |
| Context window | 1M | 1.1M |
| Input / output per 1M | $5 / $25 | $2.5 / $15 |
| Max output | 128K | 128K |
| Input | text, vision | text, vision |
| Released | 2026-04-16 | 2026-03-05 |
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 | Claude Opus 4.7 | GPT-5.4 |
|---|---|---|
| Coding | 32.2 Claude Opus 4.7 | 25.9 GPT-5.4 |
| Agents | 61.1 Claude Opus 4.7 | 62.0 GPT-5.4 |
| Reasoning | 53.6 Claude Opus 4.7 | 61.4 GPT-5.4 |
| Knowledge & language | 49.9 Claude Opus 4.7 | 46.9 GPT-5.4 |
| Human preference | 55.7 Claude Opus 4.7 | 53.1 GPT-5.4 |
Benchmarks both report
They split the 33 benchmarks both report evenly. Higher is better. Each score links to where it was published.
| Benchmark | Claude Opus 4.7 | GPT-5.4 |
|---|---|---|
| LMArena Elo | 1501 | 1475 |
| LMArena WebDev (LMArena) | 1557 | 1462 |
| MirrorCode (Epoch AI) | 43.3% | 23.3% |
| LMArena Vision (LMArena) | 1300 | 1282 |
| Chess Puzzles (Epoch AI) | 41% | 48% |
| APEX-Agents (Mercor) | 49.2% | 52.4% |
| EBR-bench (Epoch AI) | 19% | 25.4% |
| FrontierMath Tier 4 v2 (Epoch AI) | 31.7% | 49% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 70.2% | 78.6% |
| Furniture Assembly (Epoch AI) | 33.3% | 37.5% |
| GPQA Diamond | 94.2% | 92.8% |
| LiveBench Agentic Coding (LiveBench) | 50.7% | 53.8% |
| LiveBench Coding (LiveBench) | 82.1% | 77.5% |
| LiveBench Data Analysis (LiveBench) | 78.3% | 79.3% |
| LiveBench Instruction Following (LiveBench) | 66.7% | 70.2% |
| LiveBench Language (LiveBench) | 77.9% | 82.6% |
| LiveBench Mathematics (LiveBench) | 92.9% | 94.1% |
| LiveBench Reasoning (LiveBench) | 87.2% | 88.1% |
| Mystery Game Puzzles (Epoch AI) | 28% | 37% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 97.8% | 97.8% |
| SAGE (Vals AI) | 56.1% | 43.3% |
| SimpleQA Verified | 51.7% | 45.1% |
| SWE-bench Verified (Epoch AI) | 83.5% | 76.9% |
| Vending-Bench 2 (Andon Labs) | 10936.76 | 6144.18 |
| WeirdML (Håvard Tveit Ihle) | 76.4% | 77.7% |
| AIME 2026 | 95.8% | 99.2% |
| BrowseComp | 79.8% | 82.7% |
| Humanity's Last Exam (no tools) | 46.9% | 39.8% |
| Humanity's Last Exam (with tools) | 54.7% | 52.1% |
| OSWorld-Verified | 82.8% | 75% |
| SWE-Bench Pro | 64.3% | 57.7% |
| tau2-bench Banking Knowledge (Sierra) | 40.2% | 39.4% |
| GSO Opt@1 (GSO) | 42.2% | 30.4% |
FAQ
Is Claude Opus 4.7 better than GPT-5.4?
Claude Opus 4.7 scores 50.1 on the Respan Index (#42), ahead of GPT-5.4 at 50.0 (#43). They split the 33 benchmarks both report evenly.
Which is cheaper, Claude Opus 4.7 or GPT-5.4?
Claude Opus 4.7 lists at $5 input and $25 output per 1M tokens; GPT-5.4 at $2.5 and $15. GPT-5.4 is 1.8x cheaper per token at list price.
Which has the larger context window?
Claude Opus 4.7 accepts 1M tokens and GPT-5.4 1.1M tokens.
Compare other models
- vsClaude Opus 4.7 vs Claude Opus 4.6
- vsClaude Opus 4.8 vs Claude Opus 4.7
- vsGPT-5.4 vs GPT-5.2
- vsGPT-5.5 vs GPT-5.4
- vsGPT-6.1 Sol vs GPT-5.4
- vsClaude Opus 4.7 vs GPT-5.4 Pro
- vsClaude Opus 4.7 vs MiMo-V2.6-Flash
- vsClaude Opus 4.7 vs DeepSeek V4 Pro
- vsGPT-5.4 vs DeepSeek V4 Pro
- vsGPT-5.4 vs GPT-5.4 Pro
- vsGPT-5.4 vs GPT-6 Luna