OpenAI · Released Jul 9, 2026GPT-5.6 Sol
Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows.
- Capability
- 161.8ECI · #8 of 148
- Input
- $4.00per 1M tokens
- Output
- $20.00per 1M tokens
- Context
- 1.05M128K max output
What it costs
- Input
- $4.00per million tokens
- Output
- $20.00per million tokens
- Cached input
- $0.40per million tokens
- Cache write
- $5.00per million tokens
- Over 272K tokens
- $8.00 / $30.00input / output per million
- Blended (3:1)
- $8.00pricier than 92% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $18.00 |
| Team assistant25M input + 5M output tokens | $200.00 |
| Production app250M input + 50M output tokens | $2,000 |
Official OpenAI API price, as listed on models.dev.
Independent benchmarks
#8 of 148 scored models
The shaded band is Epoch AI’s confidence range (159.3–165.3); the tick marks the median scored model.
Scores from Epoch AI, run independently of OpenAI.
The details
- Lab
- OpenAI
- Released
- Jul 9, 2026
- Knowledge cutoff
- Feb 16, 2026
- Context window
- 1,050,000 tokens
- Max output
- 128,000 tokens
- Inputs
- Text, Images, PDFs
- Output
- Text
- Reasoning
- Adjustable effort low · medium · high · xhigh · max
- Tool calling
- Yes
- Structured output
- Yes
- Weights
- Proprietary
- API model ID
gpt-5.6-solon OpenAI- Availability
- 40 API providerslisted on models.dev
What OpenAI claims
Published by the lab at launch. Settings vary, so compare these only with care.
| Benchmark | Score | Setting | Source |
|---|---|---|---|
| SWE-Bench Pro | 64.6 resolve rate | — | Source |
| Terminal-Bench v2.1 | 88.8 success rate | — | Source |
| DeepSWE v1.1 | 72.7 resolve rate | — | Source |
| GPQA Diamond | 94.6 | — | Source |
| FrontierMath vv2 | 89 | — | Source |
| BrowseComp | 90.4 | — | Source |
| OSWorld v2.0 | 62.6 success rate | — | Source |
| MMMU Pro | 83 | no tools | Source |
| Agents' Last Exam | 52.7 | — | Source |
| Toolathlon | 58 success rate | — | Source |
| Artificial Analysis Intelligence Index v4.1 | 58.9 | max | Source |
| Artificial Analysis Coding Agent Index v1.1 | 80 | max | Source |
More from OpenAI
About GPT-5.6 Sol
How much does GPT-5.6 Sol cost?
GPT-5.6 Sol costs $4.00 input / $20.00 output per million tokens (official OpenAI API price). Cached input is $0.40 per million tokens, and writing to the cache costs $5.00. Requests over 272K tokens are billed at $8.00 input / $30.00 output. At a 3:1 input-to-output mix that is $8.00 per million tokens, more expensive than 92% of the 360 priced models we track.
What is the context window of GPT-5.6 Sol?
GPT-5.6 Sol accepts up to 1,050,000 tokens per request and can write up to 128,000 tokens in one response.
How good is GPT-5.6 Sol?
Epoch AI gives GPT-5.6 Sol a Capabilities Index score of 161.8 (likely range 159.3–165.3), ranking it #8 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 93.5%, FrontierMath Tiers 1–3 89.1%, OTIS Mock AIME 2024–2025 100%, SimpleQA Verified 69.7%.
Is GPT-5.6 Sol open source?
No. GPT-5.6 Sol is proprietary; you use it through OpenAI’s API or partner platforms.
What inputs does GPT-5.6 Sol support?
GPT-5.6 Sol accepts text, images and PDFs and replies in text. It is a reasoning model with low, medium, high, xhigh and max effort settings, supports tool calling and can return structured JSON output.
When was GPT-5.6 Sol released?
OpenAI released GPT-5.6 Sol on Jul 9, 2026. Its training data runs to Feb 16, 2026.
What are the best alternatives to GPT-5.6 Sol?
The closest current models from other labs on capability, price and release date are Claude Opus 5 (Anthropic, ECI 162.9, $5.00 / $25.00), Kimi K3 (Moonshot AI, ECI 157.6, $3.00 / $15.00), Grok 4.7 (xAI, $2.00 / $6.00) and Qwen3.8 Max (Alibaba (Qwen), ECI 156.6, $2.00 / $6.00).