GPT-5.6 Sol
openai/gpt-5.6-sol-20260709
OpenAI's GPT-5.6 Sol, released on 2026-07-09, accepts file, image, and text inputs and produces text output. It supports tool usage and reasoning, with a context length of 1,050,000 tokens. Pricing is 5 per
- tool use
- structured outputs
- reasoning
At a glance
Specifications
What the API gives you
- Context window
- 1.1M tokens
- Max output
- 128K tokens
- Input price
- $2 / 1M tokens
- Output price
- $10 / 1M tokens
- Cache read
- $0.20 / 1M tokens
- Modalities
- file, image, text → text
- Knowledge cutoff
- Feb 16, 2026
- Released
- Jul 9, 2026
Measured evidence
Public benchmark record
- Frontier-Bench‡
- 34.6% ± 1.62026-07-13Codex, max effort
- Frontier-Bench‡
- 37.3%2026-06-26Codex, max effort
- LiveBench Agentic Coding
- 56.2%2026-06-25
- LiveBench Coding
- 83.9%2026-06-25
- LiveBench Data Analysis
- 79.8%2026-06-25
- LiveBench Instruction Following
- 71.8%2026-06-25
- LiveBench Language
- 87.7%2026-06-25
- LiveBench Math
- 96.2%2026-06-25
- LiveBench Reasoning
- 91.7%2026-06-25
- LMArena Agent
- 10.12026-07-27
- LMArena Agent
- 10.12026-07-21
- LMArena Agent
- 102026-07-28
- LMArena Agent
- 10.12026-08-03
- LMArena Agent
- 10.82026-08-12
- LMArena Agent
- 10.32026-08-06
- LMArena Agent
- 10.72026-08-11
- LMArena Agent
- 10.92026-08-13
- LMArena Agent
- 10.32026-08-24
- LMArena Agent
- 9.82026-08-31
- LMArena Agent
- 9.32026-09-05
- LMArena Agent
- 10.22026-08-27
- LMArena Agent
- 7.52026-09-10
- LMArena Agent
- 7.42026-09-14
- LMArena Agent
- 7.12026-09-15
- LMArena Creative Writing
- 1449 Elo2026-07-21
- LMArena Creative Writing
- 1453 Elo2026-07-27
- LMArena Creative Writing
- 1453 Elo2026-07-30
- LMArena Creative Writing
- 1453 Elo2026-08-02
- LMArena Creative Writing
- 1453 Elo2026-08-03
- LMArena Creative Writing
- 1455 Elo2026-08-06
- LMArena Creative Writing
- 1455 Elo2026-08-12
- LMArena Creative Writing
- 1456 Elo2026-08-21
- LMArena Creative Writing
- 1456 Elo2026-08-27
- LMArena Creative Writing
- 1456 Elo2026-09-01
- LMArena Creative Writing
- 1456 Elo2026-09-02
- LMArena Creative Writing
- 1451 Elo2026-09-11
- LMArena Creative Writing
- 1452 Elo2026-09-13
- LMArena Instruction Following
- 1476 Elo2026-07-21
- LMArena Instruction Following
- 1473 Elo2026-07-27
- LMArena Instruction Following
- 1473 Elo2026-07-30
- LMArena Instruction Following
- 1472 Elo2026-08-02
- LMArena Instruction Following
- 1472 Elo2026-08-03
- LMArena Instruction Following
- 1471 Elo2026-08-06
- LMArena Instruction Following
- 1472 Elo2026-08-12
- LMArena Instruction Following
- 1476 Elo2026-08-21
- LMArena Instruction Following
- 1476 Elo2026-08-27
- LMArena Instruction Following
- 1477 Elo2026-09-01
- LMArena Instruction Following
- 1476 Elo2026-09-02
- LMArena Instruction Following
- 1476 Elo2026-09-11
- LMArena Instruction Following
- 1477 Elo2026-09-13
- LMArena Long Query
- 1470 Elo2026-07-21
- LMArena Long Query
- 1472 Elo2026-07-27
- LMArena Long Query
- 1472 Elo2026-07-30
- LMArena Long Query
- 1472 Elo2026-08-02
- LMArena Long Query
- 1472 Elo2026-08-03
- LMArena Long Query
- 1472 Elo2026-08-06
- LMArena Long Query
- 1473 Elo2026-08-12
- LMArena Long Query
- 1477 Elo2026-08-21
- LMArena Long Query
- 1478 Elo2026-08-27
- LMArena Long Query
- 1477 Elo2026-09-01
- LMArena Long Query
- 1478 Elo2026-09-02
- LMArena Long Query
- 1478 Elo2026-09-11
- LMArena Long Query
- 1479 Elo2026-09-13
- LMArena Math
- 1457 Elo2026-07-21
- LMArena Math
- 1453 Elo2026-07-27
- LMArena Math
- 1453 Elo2026-07-30
- LMArena Math
- 1460 Elo2026-08-02
- LMArena Math
- 1460 Elo2026-08-03
- LMArena Math
- 1459 Elo2026-08-06
- LMArena Math
- 1454 Elo2026-08-12
- LMArena Math
- 1463 Elo2026-08-21
- LMArena Math
- 1464 Elo2026-09-01
- LMArena Math
- 1461 Elo2026-08-27
- LMArena Math
- 1462 Elo2026-09-02
- LMArena Math
- 1467 Elo2026-09-11
- LMArena Math
- 1466 Elo2026-09-13
- LMArena Text
- 1485 Elo2026-07-21
- LMArena Text
- 1485 Elo2026-07-27
- LMArena Text
- 1485 Elo2026-07-30
- LMArena Text
- 1483 Elo2026-08-02
- LMArena Text
- 1482 Elo2026-08-06
- LMArena Text
- 1483 Elo2026-08-03
- LMArena Text
- 1481 Elo2026-08-10
- LMArena Text
- 1482 Elo2026-08-21
- LMArena Text
- 1481 Elo2026-08-12
- LMArena Text
- 1483 Elo2026-09-01
- LMArena Text
- 1482 Elo2026-08-27
- LMArena Text
- 1483 Elo2026-09-02
- LMArena Text
- 1482 Elo2026-09-11
- LMArena Text
- 1483 Elo2026-09-13
- LMArena Vision
- 1279 Elo2026-07-30
- LMArena Vision
- 1280 Elo2026-08-02
- LMArena Vision
- 1280 Elo2026-08-06
- LMArena Vision
- 1281 Elo2026-08-21
- LMArena Vision
- 1282 Elo2026-08-27
- LMArena Vision
- 1286 Elo2026-09-13
- LMArena WebDev
- 1630 Elo2026-07-21
- LMArena WebDev
- 1623 Elo2026-07-27
- LMArena WebDev
- 1625 Elo2026-07-24
- LMArena WebDev
- 1623 Elo2026-07-28
- LMArena WebDev
- 1623 Elo2026-07-30
- LMArena WebDev
- 1620 Elo2026-08-01
- LMArena WebDev
- 1621 Elo2026-07-31
- LMArena WebDev
- 1620 Elo2026-08-06
- LMArena WebDev
- 1620 Elo2026-08-02
- LMArena WebDev
- 1623 Elo2026-08-10
- LMArena WebDev
- 1622 Elo2026-08-12
- LMArena WebDev
- 1622 Elo2026-08-14
- LMArena WebDev
- 1616 Elo2026-09-01
- LMArena WebDev
- 1617 Elo2026-09-05
- LMArena WebDev
- 1619 Elo2026-08-21
- LMArena WebDev
- 1616 Elo2026-08-30
- LMArena WebDev
- 1617 Elo2026-09-11
- LMArena WebDev
- 1617 Elo2026-09-22
- LMArena WebDev
- 1617 Elo2026-09-23
- OSWorld Verified‡
- 28.1%2026-06-24max effort
- SWE-rebench‡
- 62.3% ± 1.82026-07-01tools
- Terminal-Bench 2.1‡
- 37.3%2026-06-26max effort
LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference. ‡ Submitted scaffold plus model result. Compare the harness and effort settings, not only the model name.
Task evidence
How broad is the support?
Each status is exact to this model version and the selected task. Official release claims remain provisional.
- Best LLM for Competition MathEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Business WritingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Logic & Problem SolvingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Marketing CopyEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Autonomous AgentsEstablished evidence
100% of intended weighted task evidence across 6 independent sources. 7 of 7 configured evaluation feeds currently have results.
- Best LLM for Chat & RoleplayEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Creative WritingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for WritingEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Roleplay & CharacterEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Math & ReasoningEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for a General AssistantEstablished evidence
100% of intended weighted task evidence across 2 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Computer & Browser UseEstablished evidence
90% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Agentic CodingEstablished evidence
82% of intended weighted task evidence across 5 independent sources. 8 of 8 configured evaluation feeds currently have results.
- Best LLM for Code CompletionEstablished evidence
81% of intended weighted task evidence across 2 independent sources. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Debugging & Code ReviewEstablished evidence
70% of intended weighted task evidence across 3 independent sources. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for CodingEstablished evidence
70% of intended weighted task evidence across 4 independent sources. 8 of 8 configured evaluation feeds currently have results.
- Best LLM for AI AgentsEstablished evidence
70% of intended weighted task evidence across 5 independent sources. 7 of 7 configured evaluation feeds currently have results.
- Best Open-Weight LLM for Self-HostingEstablished evidence
67% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Image UnderstandingEstablished evidence
65% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
- Best Local LLM for CodingEstablished evidence
64% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Fast SummariesEstablished evidence
64% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
- Best Value LLMEstablished evidence
64% of intended weighted task evidence across 2 independent sources. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Hermes AgentEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for OpenHandsEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for GooseEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for OpenClawEstablished evidence
62% of intended weighted task evidence across 3 independent sources. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Document SummarizationEstablished evidence
60% of intended weighted task evidence across 2 independent sources. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for SummarizationEstablished evidence
60% of intended weighted task evidence across 2 independent sources. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for Literary TranslationProvisional evidence
55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from WMT translation. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Everyday TranslationProvisional evidence
55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from WMT translation. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for TranslationProvisional evidence
55% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from WMT translation. 4 of 5 configured evaluation feeds currently have results.
- Best Cheap LLM for High VolumeProvisional evidence
50% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Cheapest Capable LLMProvisional evidence
50% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Massive ContextProvisional evidence
50% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LongBench v2. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for RAG & Long ContextProvisional evidence
47% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LongBench v2. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for OCR & DocumentsProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Vision & DocumentsProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Charts & DiagramsProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Document ParsingProvisional evidence
45% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Structured-output eval or VLMEvalKit tasks. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Frontend & UIProvisional evidence
40% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Design Arena Coding or Design Arena UI. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for JSON & Schema OutputProvisional evidence
39% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Structured Data ExtractionProvisional evidence
39% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Tool & Function CallingProvisional evidence
35% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or Structured-output eval. 4 of 5 configured evaluation feeds currently have results. Official release evidence is also tracked provisionally.
- Best LLM for High-Volume ExtractionProvisional evidence
31% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Route reliability or Structured-output eval. 2 of 4 configured evaluation feeds currently have results.
- Best LLM for Grounded RAGProvisional evidence
30% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from CRAG or FACTS Grounding. 2 of 5 configured evaluation feeds currently have results.
Official release evidence · provisional
What the launch material actually says
Verbatim claims from the model provider. These help on launch day but do not count as independent confirmation.
“On GeneBench v1 , which evaluates long-horizon genomics and quantitative-biology analyses, it achieves stronger results than GPT‑5.5 while using fewer tokens.”
OpenAI · Previewing GPT-5.6 Sol: a next-generation model | OpenAI“On ExploitGym (opens in a new window) 3 , a benchmark created by UC Berkeley researchers in collaboration with OpenAI and other frontier labs, GPT‑5.6 Sol, Terra, and Luna models all demonstrate strong improvements in cyber capabilitie…”
OpenAI · Previewing GPT-5.6 Sol: a next-generation model | OpenAI“On ExploitBench ², GPT‑5.6 Sol is competitive with Mythos Preview using only ~1/3 of the output tokens.”
OpenAI · Previewing GPT-5.6 Sol: a next-generation model | OpenAI“For coding workflows, GPT‑5.6 Sol sets a new state of the art on Terminal‑Bench 2.1 , which tests command-line workflows requiring planning, iteration, and tool coordination.”
OpenAI · Previewing GPT-5.6 Sol: a next-generation model | OpenAI
Recommendation coverage
Where this model appears
- #1Best LLM for Logic & Problem Solving
- #1Best LLM for Massive Context
- #1Best LLM for OpenClaw
- #1Best LLM for a General Assistant
- #1Best LLM for AI Agents
- #1Best LLM for Grounded RAG
- #1Best Cheap LLM for High Volume
- #1Best LLM for Hermes Agent
- #1Best LLM for Image Understanding
- #2Best LLM for Code Completion
- #2Best LLM for Vision & Documents
- #2Best LLM for RAG & Long Context
- #2Best LLM for Literary Translation
- #2Best LLM for Summarization
- #2Best LLM for Structured Data Extraction
- #2Best LLM for Roleplay & Character
- #2Best LLM for Math & Reasoning
- #2Best LLM for High-Volume Extraction
- #2Best LLM for Business Writing
- #2Best LLM for Fast Summaries
- #2Best LLM for Computer & Browser Use
- #2Best LLM for Agentic Coding
- #2Best LLM for Competition Math
- #2Best LLM for Chat & Roleplay
- #2Best LLM for Everyday Translation
- #2Best LLM for JSON & Schema Output
- #3Best LLM for Translation
- #3Best LLM for Marketing Copy
- #3Best LLM for Document Parsing
- #4Best Value LLM
- #4Best LLM for Coding
- #5Best LLM for Tool & Function Calling
Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.