Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 Preview
GPT-5.6 Sol
GPT-5.6 Sol is a proprietary model from OpenAI, released Jul 9, 2026. It ranks #43 of 351 on the LMArena text leaderboard with a rating of 1457. It costs $4 per million input tokens and $20 per million output tokens, with a 1.1M-token context window.
- Developer
- OpenAI
- Released
- Jul 9, 2026
- Context window
- 1.1M
- Price in / out, per 1M tokens
- $4 / $20
- Weights
- Proprietary
- Inputs
- Files, Images, Text
- Reasoning
- Yes
- Tool use
- Yes
Benchmarks
| Benchmark | Score bar | Accuracy | Rank | Cost / test |
|---|---|---|---|---|
| Terminal-Bench 4.0Coding | 37.88% | 15 / 55 | $7.98 | |
| SWE-benchCoding | 96.20% | 3 / 85 | $1.15 | |
| Vibe Code BenchCoding | 80.50% | 26 / 115 | $33.40 | |
| LiveCodeBenchCoding | 82.60% | 52 / 143 | $4 / $20 | |
| Tax Agent BenchFinance | 31.09% | 25 / 76 | $7.51 | |
| Legal Research BenchLaw | 48.08% | 11 / 85 | $21.61 | |
| MedCodeMedicine | 43.97% | 43 / 107 | $4 / $20 | |
| GPQA DiamondReasoning | 95.20% | 2 / 138 | $4 / $20 | |
| MMLU-ProReasoning | 89.10% | 16 / 138 | $4 / $20 | |
| MMMUReasoning | 88.84% | 6 / 93 | $4 / $20 | |
| Terminal-Bench ScienceReasoning | 20.00% | 11 / 42 | $7.32 |
Vals AI, as of Oct 24, 2026.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1457 | 43 / 351 | 42.7K |
| CodingLMArena | 1496 | 31 / 346 | 11.4K |
| MathLMArena | 1473 | 43 / 339 | 2K |
| Creative writingLMArena | 1450 | 34 / 349 | 9K |
| Instruction followingLMArena | 1483 | 17 / 351 | 14.9K |
| Hard promptsLMArena | 1484 | 32 / 351 | 27.7K |
| Multi-turnLMArena | 1461 | 46 / 349 | 7K |
| Long queriesLMArena | 1481 | 28 / 330 | 19.7K |
| ExpertLMArena | 1516 | 15 / 305 | 4.8K |
| Legal and governmentIndustries | 1475 | 36 / 325 | 3.6K |
| Medicine and healthcareIndustries | 1431 | 114 / 325 | 3.1K |
| Business and financeIndustries | 1457 | 38 / 344 | 7.8K |
| ChineseLanguages | 1526 | 27 / 328 | 3.3K |
| SpanishLanguages | 1442 | 77 / 249 | 1.3K |
| GermanLanguages | 1475 | 25 / 261 | 674 |
| FrenchLanguages | 1472 | 45 / 252 | 1.4K |
| JapaneseLanguages | 1473 | 21 / 236 | 681 |
| WebDevLMArena | 1615 | 29 / 124 | 18.2K |
| VisionLMArena | 1281 | 35 / 146 | 11.4K |
LMArena, CC BY 4.0, as of Oct 22, 2026. Where a model is rated at several effort levels, its best is shown (GPT-5.6 Sol (xHigh), GPT-5.6 Sol (xHigh, Codex Harness)).
Compare GPT-5.6 Sol
OpenAI
The lab behind GPT and ChatGPT, which brought large language models to hundreds of millions of people. Founded 2015, based in San Francisco. openai.com
Questions
Who makes GPT-5.6 Sol?
GPT-5.6 Sol is made by OpenAI, founded 2015 and based in San Francisco.
How much does GPT-5.6 Sol cost?
$4 per million input tokens and $20 per million output tokens, at its provider’s list price.
What is GPT-5.6 Sol’s context window?
1.1M tokens (1,050,000).
How does GPT-5.6 Sol rank?
#43 of 351 on LMArena’s text leaderboard, rated 1457 (95% interval 1453–1462) from 42,713 votes, as of Oct 22, 2026.
Is GPT-5.6 Sol open source?
No. GPT-5.6 Sol is proprietary and served through its provider’s API.