GPT-5.5 vs GPT-5.6 Sol: API Pricing & Benchmarks
GPT-5.5 vs GPT-5.6 Sol — or, searched the other way round, GPT-5.6 Sol vs GPT-5.5: official API prices, what the same workload costs on each, and every published benchmark score, side by side. Every figure comes straight from the data below; this page does not pick a winner.
Price per 1M tokens
| GPT-5.5 | GPT-5.6 Sol | |
|---|---|---|
| Input | $5.00 | $4.00 |
| Output | $30.00 | $20.00 |
| Blended | $11.25 | $8.00 |
| Developer | OpenAI | OpenAI |
| Released | 2026-04-23 | 2026-07-09 |
GPT-5.5: Higher rates apply above a context threshold; see the official site
GPT-5.6 Sol: Higher rates apply above a context threshold; see the official site
Prices: official list prices as recorded by models.dev (GPT-5.5) and models.dev (GPT-5.6 Sol).
Cost of the same workload
| Workload | Tokens | GPT-5.5 | GPT-5.6 Sol | Difference |
|---|---|---|---|---|
| Short chat reply | 500 in / 300 out | $0.0115 | 0.8¢ | GPT-5.6 Sol cheaper by 0.35¢ |
| RAG query | 4,000 in / 500 out | $0.0350 | $0.0260 | GPT-5.6 Sol cheaper by 0.9¢ |
| Long document summary | 50,000 in / 1,000 out | $0.2800 | $0.2200 | GPT-5.6 Sol cheaper by $0.0600 |
| Code review of a 2k-line file | 30,000 in / 2,000 out | $0.2100 | $0.1600 | GPT-5.6 Sol cheaper by $0.0500 |
| 1,000 chat replies/day for 30 days | 30,000 × (500 in / 300 out) | $345.00 | $240.00 | GPT-5.6 Sol cheaper by $105.00 |
Cost = (input tokens ÷ 1,000,000 × input price) + (output tokens ÷ 1,000,000 × output price), using each model's own prices above.
Benchmarks
| Benchmark | GPT-5.5 | GPT-5.6 Sol |
|---|---|---|
| Epoch Capabilities Index (ECI) | 160.0 epoch · as of 2026-04-23 |
163.0 epoch · as of 2026-07-09 |
| Epoch Mock AIME accuracy | 84.4% epoch · as of 2026-04-23 |
100.0% epoch · as of 2026-07-09 |
| Epoch FrontierMath (T1–3) accuracy | 85.3% epoch · as of 2026-04-23 |
89.1% epoch · as of 2026-07-09 |
| Arena text rating | 1477 arena · as of 2026-10-28 |
1485 arena · as of 2026-10-28 |
| Arena coding rating (human preference votes, not task accuracy) | 1510 arena · as of 2026-10-28 |
1533 arena · as of 2026-10-28 |
| Arena math rating | 1500 arena · as of 2026-10-28 |
1495 arena · as of 2026-10-28 |
Full pages
GPT-5.5 pricing GPT-5.6 Sol pricing GPT-5.5 token counter GPT-5.6 Sol token counter
FAQ
Which is cheaper, GPT-5.5 or GPT-5.6 Sol?
GPT-5.6 Sol is cheaper on blended price: $8.00 vs $11.25 per 1M tokens, which is $3.25 less per 1M tokens (28.9% below GPT-5.5).
How do the input and output prices of GPT-5.5 and GPT-5.6 Sol compare?
GPT-5.5 costs $5.00 per 1M input tokens and $30.00 per 1M output tokens. GPT-5.6 Sol costs $4.00 per 1M input tokens and $20.00 per 1M output tokens.
What are the ECI scores of GPT-5.5 and GPT-5.6 Sol?
GPT-5.5 scores 160.0 on the Epoch Capabilities Index (ECI); GPT-5.6 Sol scores 163.0 on the Epoch Capabilities Index (ECI).
Where does the GPT-5.5 vs GPT-5.6 Sol comparison data come from?
Prices come from models.dev and LiteLLM (open MIT-licensed catalogs); benchmark scores come from Epoch AI and LMArena (open CC BY 4.0 datasets). This page states the numbers side by side and does not pick an overall winner — see /data-sources/.