GPT-5.6 Sol
- Context
- 1.05M
- 128K max output
- Input price
- $4.00
- per 1M tokens · OpenAI API
- Output price
- $20.00
- per 1M tokens
- Epoch Capabilities Index
- 162
- #6 of 71 listed
- LMArena Text
- 1,483
- #14 of 79 listed
Where to run it
3 offerings · prices as listed| Meter | Tier | Price |
|---|---|---|
| Input | Standard | $4.00 / 1M tokens |
| Batch | $2.00 / 1M tokens | |
| Flex tier | $2.00 / 1M tokens | |
| Fast tier | $8.00 / 1M tokens | |
| Output | Standard | $20.00 / 1M tokens |
| Batch | $10.00 / 1M tokens | |
| Flex tier | $10.00 / 1M tokens | |
| Fast tier | $40.00 / 1M tokens | |
| Cache read | Standard | $0.40 / 1M tokens |
| Batch | $0.20 / 1M tokens | |
| Flex tier | $0.20 / 1M tokens | |
| Fast tier | $0.80 / 1M tokens | |
| Cache write | Standard | $5.00 / 1M tokens |
| Batch | $2.50 / 1M tokens | |
| Flex tier | $2.50 / 1M tokens | |
| Fast tier | $10.00 / 1M tokens | |
Azure OpenAI in Foundry (opens in a new tab) gpt-5-6-sol | ||
| Input | Standard | $4.00 / 1M tokens |
| Priority tier | $8.00 / 1M tokens | |
| Output | Standard | $20.00 / 1M tokens |
| Priority tier | $40.00 / 1M tokens | |
| Cache read | Standard | $0.40 / 1M tokens |
| Priority tier | $0.80 / 1M tokens | |
| Cache write | Standard | $5.00 / 1M tokens |
| Priority tier | $10.00 / 1M tokens | |
Azure OpenAI in Foundry (opens in a new tab) gpt-5-6-sol | ||
| Input | Standard | $4.40 / 1M tokens |
| Priority tier | $8.80 / 1M tokens | |
| Output | Standard | $22.00 / 1M tokens |
| Priority tier | $44.00 / 1M tokens | |
| Cache read | Standard | $0.44 / 1M tokens |
| Priority tier | $0.88 / 1M tokens | |
| Cache write | Standard | $5.50 / 1M tokens |
| Priority tier | $11.00 / 1M tokens | |
Changes
All changes →- 8 prices dropped on OpenAI API Input (Prompts over 272K tokens) $8.00 · Output (Prompts over 272K tokens) $30.00 · Cache read (Prompts over 272K tokens) $0.80 · Input (Prompts over 272K tokens, Fast tier) $16.00 · +4 more
Vendor-reported scores
Published by OpenAI, under the conditions it states| Benchmark | Setting | Score |
|---|---|---|
| SWE-bench Pro | — | 64.6% |
| DeepSWE 1.1 | — | 72.7% |
| Terminal-Bench 2.1 | single agent | 88.8% |
| ultra (multi-agent, 4 agents) | 91.9% | |
| OSWorld 2.0 | set and scoring not stated | 62.6% |
| v2026.08.08 offline set, partial score; maximum across effort levels | 65.7% | |
| BrowseComp | single agent | 90.4% |
| ultra (multi-agent, 4 agents) | 92.2% | |
| GPQA Diamond | — | 94.6% |
| MMMU-Pro | no tools | 83% |
| with tools | 84.6% | |
| MRCR v2 | 8-needle, 256K–512K | 91.5% |
| 8-needle, 512K–1M | 73.8% | |
| Terminal-Bench 4.0 | maximum across effort levels | 37.3% |
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 162
#6 of 71 listed models · CI 159.6–165.9
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text 1,483
#14 of 79 listed models · xhigh effort · CI 1,479–1,488 · 34,258 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- Epoch AI · FrontierMath Tiers 1–3 89.1%
#3 of 56 listed models · max effort · ±1.85
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · GPQA Diamond 93.5%
#9 of 64 listed models · max effort · ±1.57
Also listed: low effort 89.9% · reasoning off 82.8%
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text · Chinese 1,538
#8 of 79 listed models · xhigh effort · CI 1,526–1,550 · 2,675 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Text · Coding 1,531
#14 of 79 listed models · xhigh effort · CI 1,524–1,538 · 9,391 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Vision 1,280
#20 of 49 listed models · xhigh effort · CI 1,272–1,289 · 7,729 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 13, 2026 · changes we made