Claude Sonnet 5 vs GPT-6 Luna
| Spec | Claude Sonnet 5 | GPT-6 Luna |
|---|---|---|
| Respan Index | 48.0 (#47) | 48.6 (#45) |
| Blended price per 1M | $4 | $0.2 |
| Context window | 1M | 1.1M |
| Input / output per 1M | $2 / $10 | $0.1 / $0.5 |
| Max output | 128K | 128K |
| Input | text, vision | text, vision |
| Released | 2026-06-30 | 2026-09-22 |
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 Sonnet 5 | GPT-6 Luna |
|---|---|---|
| Coding | 32.9 Claude Sonnet 5 | 28.4 GPT-6 Luna |
| Agents | 62.1 Claude Sonnet 5 | 59.9 GPT-6 Luna |
| Reasoning | 50.6 Claude Sonnet 5 | 57.1 GPT-6 Luna |
| Knowledge & language | 34.1 Claude Sonnet 5 | 38.9 GPT-6 Luna |
| Human preference | 54.9 Claude Sonnet 5 | 55.6 GPT-6 Luna |
Benchmarks both report
Claude Sonnet 5 leads on 14 of the 22 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Claude Sonnet 5 | GPT-6 Luna |
|---|---|---|
| LMArena Agent (LMArena) | 0.01 | 0.0064 |
| LMArena Elo | 1461 | 1441 |
| LMArena WebDev (LMArena) | 1539 | 1578 |
| MirrorCode (Epoch AI) | 63.3% | 26.7% |
| LMArena Vision (LMArena) | 1261 | 1235 |
| Chess Puzzles (Epoch AI) | 45% | 36% |
| Terminal-Bench 4.0 (Vals AI) | 10.6% | 13.6% |
| APEX-Agents (Mercor) | 54.5% | 44.3% |
| FrontierMath Tier 4 v2 (Epoch AI) | 29.3% | 56.1% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 65.6% | 78.9% |
| GPQA Diamond | 90.5% | 90.5% |
| LiveBench Agentic Coding (LiveBench) | 59.4% | 51.2% |
| LiveBench Coding (LiveBench) | 80.7% | 79% |
| LiveBench Data Analysis (LiveBench) | 71.7% | 73.4% |
| LiveBench Instruction Following (LiveBench) | 63.9% | 55.9% |
| LiveBench Language (LiveBench) | 75% | 73.8% |
| LiveBench Mathematics (LiveBench) | 92.9% | 89.1% |
| LiveBench Reasoning (LiveBench) | 88.7% | 81.8% |
| Mystery Game Puzzles (Epoch AI) | 35% | 7% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 94.7% | 98.9% |
| SAGE (Vals AI) | 48.9% | 48.1% |
| SimpleQA Verified | 33.7% | 41.4% |
FAQ
Is Claude Sonnet 5 better than GPT-6 Luna?
GPT-6 Luna scores 48.6 on the Respan Index (#45), ahead of Claude Sonnet 5 at 48.0 (#47). Claude Sonnet 5 leads on 14 of the 22 benchmarks both report.
Which is cheaper, Claude Sonnet 5 or GPT-6 Luna?
Claude Sonnet 5 lists at $2 input and $10 output per 1M tokens; GPT-6 Luna at $0.1 and $0.5. GPT-6 Luna is 20x cheaper per token at list price.
Which has the larger context window?
Claude Sonnet 5 accepts 1M tokens and GPT-6 Luna 1.1M tokens.
Compare other models
- vsClaude Sonnet 5 vs Claude Sonnet 4.6
- vsClaude Sonnet 5.5 vs Claude Sonnet 5
- vsGPT-6 Luna vs GPT-5.6 Luna
- vsGemini 3.8 Flash vs Claude Sonnet 5
- vsGPT-6 Luna vs DeepSeek V4 Flash
- vsClaude Sonnet 5 vs GPT-5.6 Luna
- vsClaude Sonnet 5 vs Hy4 preview
- vsClaude Sonnet 5 vs Muse Spark 1.1
- vsGPT-6 Luna vs DeepSeek V4 Pro
- vsGPT-6 Luna vs GPT-5.4