Compare Start here

OpenAILanguage modelSuperseded

GPT-5.6 Sol

Introduced the Sol naming and served as OpenAI's flagship through summer 2026.

Compare head-to-head Race it in 3D

Overall Intelligence

#10

of 56
3.9 pts behind #1

0

races led

22

weeks at #1

84.9

score

Released: Jun 25, 2026; 4 mo ago
Context: 1M; 128K max output
Price: $4 / $20; per 1M in/out · $8.00 blended
Speed: 100 tok/s; 5s to first token
Modalities: text · image · audio · video; outputs text
Weights: Closed; API and apps

The 30-second version

Introduced the Sol naming and served as OpenAI's flagship through summer 2026.

Key takeaways

1Introduced the Sol naming and served as OpenAI's flagship through summer 2026.

Where it ranks

Current snapshot · illustrative scores · click any race to watch it

GPT-5.6 SolBest on each axis

Overall

10Overall+3.9 pts

Coding

11Coding+6.8 pts 15Frontend & UI+51 Elo 12Backend & APIs+6.8 pts 13Distributed Systems+8.4 pts 12Full-Stack Apps+8.4 pts 11Debugging & Repair+7.3 pts 12Agentic Coding+8.8 pts 8Code Review & Security+4.6 pts 9Data Science & SQL+3.9 pts 9Mobile Apps+5.4 pts 10Terminal & DevOps+9.8 pts 10Competitive Programming+114

Hard Reasoning

8Hard Reasoning+4.4 pts 9Frontier Questions+7.3 pts 8Abstract & Novel Reasoning+6.8 pts 9Logic & Puzzles+3.0 pts 10Planning & Strategy+5.9 pts

Math

9Math+4.3 pts 5Competition Math+3.2 pts 8Research Math+5.7 pts 8Applied & Quantitative+4.0 pts

Science

11Science+6.4 pts 11Graduate Science QA+4.1 pts 14Biology & Medicine+6.8 pts 9Physics & Chemistry+3.5 pts 14Research Assistance+7.0 pts

Writing & Language

10Writing & Language+5.3 pts 11Creative Writing+39 Elo 11Technical Writing+5.2 pts 11Editing & Style+4.9 pts 10Multilingual+3.9 pts 8Instruction Following+4.2 pts 8Conversation & EQ+27 Elo

Multimodal

9Multimodal+10.1 pts 15Image Understanding+7.9 pts 15Documents & Charts+8.2 pts 10Video Understanding+12.7 pts 5Speech & Audio+5.6 pts

Tool Use & Agents

9Tool Use & Agents+6.1 pts 9Tool Use+4.6 pts 11Computer Use+9.1 pts 10Web Research+7.6 pts 10Long-Horizon+17.6 hrs

Long Context

12Long Context+6.8 pts 111M Recall+6.0 pts 12Long Reasoning+5.7 pts

Speed & Cost

28Speed & Cost+41.2 pts 22Output Speed+422 tok/s 33Latency+4.38s 31Cost-Efficiency+46.6 pts

Trust & Factuality

11Trust & Factuality+7.5 pts 12Factuality+6.6 pts 11Honesty & Calibration+5.8 pts

Overall Intelligence over time

#1 frontierGPT-5.6 SolSonnet 5.5Grok 4.7

Update history

The distilled version first; open any release for the deeper breakdown.

The GPT Sol family

GPT-5.6 SolJune 2026 GPT-6 SolSeptember 2026 GPT-6.1 SolSeptember 2026

60-second briefing

GPT-5.6 Sol, explained

AI video summary · placeholder render from the auto-written script

From OpenAI's own guide, distilled

Prompting guides for GPT-5 and later models

Reasoning models do their own step-by-step thinking; give goals and constraints rather than scripting the chain of thought.
Use the reasoning-effort and verbosity controls instead of prompt tricks to trade depth for speed.
Contradictory instructions waste reasoning tokens; audit system prompts for conflicts.
For agents, define stop conditions and how eager the model should be to keep going without asking.
Structured outputs (JSON schema) beat asking politely for a format.

All official guides, distilled →

About OpenAI

Maker of ChatGPT and the GPT series, pursuing general-purpose AI at consumer scale.

Optimising for

Frontier reasoning and math, broad consumer usefulness inside ChatGPT, and tiered models (Astra, Sol, Luna) that trade depth for price and speed.

San Francisco, US · founded 2015

Closest rivals

GPT-5.6 Sol is 10th in overall intelligence; leader is Claude Opus 5.5.