GPT-6.1 Sol vs GPT-6 Sol

The match-up at a glance

GPT-6.1 Sol and GPT-6 Sol are two frontier models, the ones you pick when quality comes first, both built by OpenAI. GPT-6.1 Sol was released on September 29, 2026, GPT-6 Sol on September 22, 2026.

All figures below come from public sources checked on October 22, 2026: models.dev and OpenRouter for prices and capabilities, Epoch AI for scores.

Which one is more capable?

On our capability index, GPT-6.1 Sol scores 92.1 out of 100 and GPT-6 Sol 89.2. GPT-6.1 Sol leads by 2.9 points.

On the 5 benchmarks reported for both, GPT-6.1 Sol wins 4 times and GPT-6 Sol 0:

• GPQA Diamond (PhD-level science questions): GPT-6.1 Sol 95.4%, GPT-6 Sol 94.3%
• OTIS Mock AIME (competition mathematics problems): GPT-6.1 Sol 100%, GPT-6 Sol 100%
• FrontierMath (very hard research-level mathematics): GPT-6.1 Sol 93.7%, GPT-6 Sol 89.8%
• FrontierCode (advanced programming tasks): GPT-6.1 Sol 50.2%, GPT-6 Sol 49.3%
• APEX-Agents (professional tasks carried out by an agent): GPT-6.1 Sol 60%, GPT-6 Sol 54.3%

Pricing

GPT-6.1 Sol costs $2 per million input tokens and $10 per million output tokens; GPT-6 Sol costs $2 and $10.

For a typical agent task (read 20,000 tokens, write 2,000), expect $0.06 with GPT-6.1 Sol versus $0.06 with GPT-6 Sol. The cost is almost the same.

Run the numbers with your own volumes in the agent cost calculator.

Features and limits

Both models have the same context window: 1,050,000 tokens, roughly 1,575 pages.

Reasoning mode: GPT-6.1 Sol yes, GPT-6 Sol yes. Tool calling: GPT-6.1 Sol yes, GPT-6 Sol yes. Image input: GPT-6.1 Sol yes, GPT-6 Sol yes.

Neither publishes its weights: both are used through an API.

Knowledge cutoff: April 30, 2026 for GPT-6.1 Sol, April 20, 2026 for GPT-6 Sol.

In short

Our reading of the data:

• Choose GPT-6.1 Sol for a higher capability index.

This verdict relies on public figures; it does not replace a trial on your own tasks.