AI models

GPT-6.1 Sol vs Gemini 3.8 Flash vs Claude Sonnet 5.5

GPT-6.1 Sol, Gemini 3.8 Flash and Claude Sonnet 5.5 on 5 benchmarks and price per 1M tokens.

GPT-6.1 Sol

US$ 2.00 / US$ 10.00 prompts up to 272,000 tokens

Input US$/1M / Output US$/1M · Official pricing

Gemini 3.8 Flash

US$ 0.75 / US$ 3.75Current price, valid through December 31, 2026 On the free tier, submitted data is used to improve Google's products Cache read US$ 0.15 Batch US$ 0.75 input and US$ 3.75 output

Input US$/1M / Output US$/1M · Official pricing

Claude Sonnet 5.5

US$ 2.00 / US$ 10.00 1 million token context at the same price · US only ×1.1

Input US$/1M / Output US$/1M · Official pricing

GPQA Diamond

95.4% · 95.4% · 95.6%

GPT-6.1 Sol · Gemini 3.8 Flash · Claude Sonnet 5.5

01

Direct answer

Direct answer: GPQA Diamond: GPT-6.1 Sol 95.4%, Gemini 3.8 Flash 95.4%, Claude Sonnet 5.5 95.6% (Epoch AI). FrontierMath (tiers 1 to 3): GPT-6.1 Sol 93.7%, Gemini 3.8 Flash 68.4%, Claude Sonnet 5.5 88.8% (Epoch AI). SimpleQA Verified: GPT-6.1 Sol 73.9%, Gemini 3.8 Flash 69.7%, Claude Sonnet 5.5 46.5% (Epoch AI). τ-bench Verified (airline): GPT-6.1 Sol 70.7%, Gemini 3.8 Flash 76.7%, Claude Sonnet 5.5 78.8% (OpenRouter Benchmarks).
see more

02

On GPQA Diamond, Claude Sonnet 5.5 is in a statistical tie with Gemini 3.8 Flash, Claude Sonnet 5.5 is in a statistical tie with GPT-6.1 Sol and Gemini 3.8 Flash is in a statistical tie with GPT-6.1 Sol. On FrontierMath, Claude Sonnet 5.5 leads Gemini 3.8 Flash, GPT-6.1 Sol leads Claude Sonnet 5.5 and GPT-6.1 Sol leads Gemini 3.8 Flash. On SimpleQA Verified, Gemini 3.8 Flash leads Claude Sonnet 5.5, GPT-6.1 Sol leads Claude Sonnet 5.5 and GPT-6.1 Sol leads Gemini 3.8 Flash. On τ-bench Verified, Claude Sonnet 5.5 is in a statistical tie with Gemini 3.8 Flash, Claude Sonnet 5.5 is in a statistical tie with GPT-6.1 Sol and Gemini 3.8 Flash is in a statistical tie with GPT-6.1 Sol.

GPT-6.1 Sol

  • FrontierMath +4.9 p.p.
  • SimpleQA Verified +4.2 p.p.

statistical tie

  • GPQA Diamond statistical tie
  • τ-bench Verified statistical tie

03

MetricGPT-6.1 SolGemini 3.8 FlashClaude Sonnet 5.5Source
Input US$/1MUS$ 2.00 / US$ 10.00 Official pricing prompts up to 272,000 tokensUS$ 0.75 / US$ 3.75 Official pricing Current price, valid through December 31, 2026 On the free tier, submitted data is used to improve Google's products Cache read US$ 0.15 Batch US$ 0.75 input and US$ 3.75 outputUS$ 2.00 / US$ 10.00 Official pricing 1 million token context at the same price · US only ×1.1
Output US$/1MUS$ 10.00US$ 3.75US$ 10.00
Context1,050,0001,048,5761,000,000
BENCHMARKS
GPQA Diamond statistical tie95.4%between 3rd and 12th of 250 statistical tie95.4%between 3rd and 12th of 250 statistical tie95.6%between 2nd and 11th of 250 statistical tieEpoch AI
OTIS Mock AIME saturated100% top score98.9% statistical tie with the top score100% top scoreEpoch AI
FrontierMath (tiers 1 to 3)93.7%between 2nd and 3rd of 85 68.4%between 32nd and 35th of 85 88.8%between 9th and 11th of 85 Epoch AI
SimpleQA Verified73.9%between 6th and 7th of 87 69.7%between 10th and 13th of 87 46.5%between 37th and 42nd of 87 Epoch AI
τ-bench Verified (airline)70.7%between 80th and 89th of 130 76.7%between 22nd and 46th of 130 78.8%between 6th and 21st of 130 OpenRouter Benchmarks
OTIS Mock AIME saturated100% top score98.9% statistical tie with the top score100% top scoreEpoch AI

Compare with

How to read

  1. Benchmark scores are percentages, and higher is better.
  2. In each row, the highlight marks the model ahead by more than the benchmark's margin; within the margin, the row shows «statistical tie».
  3. A saturated benchmark stays in the table, with its badge, and does not count toward the verdict.
  4. Prices are in US$ per 1M tokens and carry the source label: «Official lab price», «Third-party provider» or «No current price».
  5. Every number comes with its source and date. «No data» means no measurement exists in the open sources this section uses.

Sources

  • Epoch AI, CC BY 4.0, latest version we checked.
  • OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
  • OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.

How to cite

If you use this data, cite the page and the date you accessed it.

Adriano Meirinho. "GPT-6.1 Sol vs Gemini 3.8 Flash vs Claude Sonnet 5.5". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/comparar/claude-sonnet-5-5-vs-gemini-3-8-flash-vs-gpt-6-1-sol/. Accessed November 21, 2026.

Third-party data follows the license of each source. The full table is available as CSV and JSON.

Markdown