GPT-5.4 vs GPT-6 Sol
GPT-6 Sol leads GPT-5.4 by 12.0 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.4 wins 0 and GPT-6 Sol wins 17 of 17 shared benchmarks. GPT-6 Sol is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | GPT-5.4 | GPT-6 Sol |
|---|---|---|
| Agentic coding | 57 | 72 |
| Long-horizon autonomy | 61 | 76 |
| Knowledge-work agents | 40 | 48 |
| Computer use | 42 | 54 |
| Security & context learning | 54 | 59 |
Shared benchmarks
| Benchmark | GPT-5.4 | GPT-6 Sol |
|---|---|---|
| APEX-Agentsindex | 52.4% | 54.3% |
| Vending-Bench 2index | $6,144 | $14,428 |
| ARC-AGI | 93.7% | 95.5% |
| ARC-AGI-2 | 74.0% | 89.6% |
| Chess Puzzles | 48.0% | 56.0% |
| CritPt | 23.4% | 30.9% |
| DTBench | 94.4% | 97.3% |
| FrontierMath Tier 4 | 49.0% | 90.0% |
| FrontierMath Tiers 1-3 | 78.6% | 89.8% |
| GPQA Diamond | 93.3% | 94.3% |
| LMCA | 52.0% | 59.1% |
| MirrorCode | 23.3% | 48.9% |
| OTIS Mock AIME | 97.8% | 100.0% |
| ProofBench | 56.0% | 83.0% |
| SciCode | 56.6% | 57.6% |
| SimpleQA Verified | 45.1% | 60.7% |
| WebDev Arena | 1463 | 1689 |
Specs and price
| GPT-5.4 | GPT-6 Sol | |
|---|---|---|
| Agent Index | 51.6 (#33) | 63.6 (#13) |
| Agent $/M | $1.27 | $0.92 |
| Input / output $/M | $2.50 / $15.00 | $2.00 / $10.00 |
| Context | 1.1M | 1.1M |
| Released | Mar 5, 2026 | Sep 22, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 5 |
Don’t pick one. Route.
Shogo can run GPT-5.4 and GPT-6 Sol side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo