GPT-5.4 vs GPT-6.1 Sol
| Spec | GPT-5.4 | GPT-6.1 Sol |
|---|---|---|
| Respan Index | 49.5 (#33) | 69.8 (#7) |
| Blended price per 1M | $3.44 | $2 |
| Context window | 1.1M | 1.1M |
| Input / output per 1M | $1.25 / $10 | $1 / $5 |
| Max output | 128K | 128K |
| Input | text, vision | text, vision |
| Released | 2026-03-05 | 2026-09-29 |
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-5.4 | GPT-6.1 Sol |
|---|---|---|
| Coding | 24.3 GPT-5.4 | 53.3 GPT-6.1 Sol |
| Agents | 62.0 GPT-5.4 | 73.1 GPT-6.1 Sol |
| Reasoning | 61.5 GPT-5.4 | 87.2 GPT-6.1 Sol |
| Knowledge & language | 46.9 GPT-5.4 | 74.4 GPT-6.1 Sol |
| Human preference | 52.6 GPT-5.4 | 59.8 GPT-6.1 Sol |
Benchmarks both report
GPT-6.1 Sol leads on 23 of the 23 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | GPT-5.4 | GPT-6.1 Sol |
|---|---|---|
| Chess Puzzles (Epoch AI) | 48% | 68% |
| LMArena Elo | 1475 | 1481 |
| LMArena WebDev (LMArena) | 1464 | 1751 |
| LMArena Vision (LMArena) | 1284 | 1294 |
| LMArena Agent (LMArena) | 0.0267 | 0.1247 |
| APEX-Agents (Mercor) | 52.4% | 60% |
| EBR-bench (Epoch AI) | 25.4% | 54.3% |
| FrontierMath Tier 4 v2 (Epoch AI) | 49% | 100% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 78.6% | 93.7% |
| Furniture Assembly (Epoch AI) | 37.5% | 80% |
| GPQA Diamond | 92.8% | 95.4% |
| LiveBench Agentic Coding (LiveBench) | 53.8% | 56.8% |
| LiveBench Coding (LiveBench) | 77.5% | 80.7% |
| LiveBench Data Analysis (LiveBench) | 79.3% | 82.7% |
| LiveBench Instruction Following (LiveBench) | 70.2% | 74.2% |
| LiveBench Language (LiveBench) | 82.6% | 90.1% |
| LiveBench Mathematics (LiveBench) | 94.1% | 96.8% |
| LiveBench Reasoning (LiveBench) | 88.1% | 92.6% |
| MirrorCode (Epoch AI) | 15.6% | 61.1% |
| Mystery Game Puzzles (Epoch AI) | 37% | 80% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 97.8% | 100% |
| SAGE (Vals AI) | 43.3% | 46.5% |
| SimpleQA Verified | 45.1% | 73.9% |
FAQ
Is GPT-5.4 better than GPT-6.1 Sol?
GPT-6.1 Sol scores 69.8 on the Respan Index (#7), ahead of GPT-5.4 at 49.5 (#33). GPT-6.1 Sol leads on 23 of the 23 benchmarks both report.
Which is cheaper, GPT-5.4 or GPT-6.1 Sol?
GPT-5.4 lists at $1.25 input and $10 output per 1M tokens; GPT-6.1 Sol at $1 and $5. GPT-6.1 Sol is 1.7x cheaper per token at list price.
Which has the larger context window?
Both accept 1.1M tokens of context.