AI models
GPT-6.2 Astra vs Gemini 4 Carbon vs Claude Fable 5.2
GPT-6.2 Astra, Gemini 4 Carbon and Claude Fable 5.2 on 4 benchmarks and price per 1M tokens.
GPT-6.2 Astra
US$ 8.00 / US$ 40.00 prompts up to 272,000 tokensInput US$/1M / Output US$/1M · Official pricing
Gemini 4 Carbon
US$ 2.00 / US$ 10.00Current price, valid through November 30, 2026 Cache read US$ 0.20 Batch US$ 2.00 input and US$ 10.00 output submitted data not used for trainingInput US$/1M / Output US$/1M · Official pricing
Claude Fable 5.2
US$ 8.00 / US$ 40.00 1 million token context at the same price · US only ×1.1Input US$/1M / Output US$/1M · Official pricing
GPQA Diamond
96.8% · 96.5% · 93.9%
GPT-6.2 Astra · Gemini 4 Carbon · Claude Fable 5.2
01
Direct answer
see more
02
On GPQA Diamond, Claude Fable 5.2 is in a statistical tie with Gemini 4 Carbon, Claude Fable 5.2 is in a statistical tie with GPT-6.2 Astra and Gemini 4 Carbon is in a statistical tie with GPT-6.2 Astra. On FrontierMath, Claude Fable 5.2 leads Gemini 4 Carbon, Claude Fable 5.2 is in a statistical tie with GPT-6.2 Astra and Gemini 4 Carbon is in a statistical tie with GPT-6.2 Astra. On SimpleQA Verified, Claude Fable 5.2 leads Gemini 4 Carbon, Claude Fable 5.2 leads GPT-6.2 Astra and Gemini 4 Carbon is in a statistical tie with GPT-6.2 Astra.
Claude Fable 5.2
- SimpleQA Verified +15.5 p.p.
statistical tie
- GPQA Diamond statistical tie
03
| Metric | GPT-6.2 Astra | Gemini 4 Carbon | Claude Fable 5.2 | Source |
|---|---|---|---|---|
| Input US$/1M | US$ 8.00 / US$ 40.00 Official pricing prompts up to 272,000 tokens | US$ 2.00 / US$ 10.00 Official pricing submitted data not used for training Current price, valid through November 30, 2026 Cache read US$ 0.20 Batch US$ 2.00 input and US$ 10.00 output | US$ 8.00 / US$ 40.00 Official pricing 1 million token context at the same price · US only ×1.1 | |
| Output US$/1M | US$ 40.00 | US$ 10.00 | US$ 40.00 | |
| Context | 1,050,000 | 1,048,576 | 1,000,000 | |
| BENCHMARKS | ||||
| GPQA Diamond statistical tie | 96.8%between 1st and 4th of 250 statistical tie | 96.5%between 1st and 6th of 250 statistical tie | 93.9%between 9th and 28th of 250 statistical tie | Epoch AI |
| OTIS Mock AIME saturated | 100% top score | 100% top score | 99.7% statistical tie with the top score | Epoch AI |
| FrontierMath (tiers 1 to 3) | 92.6%4th of 85 | 90.5%between 5th and 9th of 85 | 95.1%1st of 85 top score | Epoch AI |
| SimpleQA Verified | 78.5%between 2nd and 3rd of 87 | 77.7%between 2nd and 3rd of 87 | 94%1st of 87 top score | Epoch AI |
| OTIS Mock AIME saturated | 100% top score | 100% top score | 99.7% statistical tie with the top score | Epoch AI |
Compare with
How to read
- Benchmark scores are percentages, and higher is better.
- In each row, the highlight marks the model ahead by more than the benchmark's margin; within the margin, the row shows «statistical tie».
- A saturated benchmark stays in the table, with its badge, and does not count toward the verdict.
- Prices are in US$ per 1M tokens and carry the source label: «Official lab price», «Third-party provider» or «No current price».
- 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-6.2 Astra vs Gemini 4 Carbon vs Claude Fable 5.2". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/comparar/claude-fable-5-2-vs-gemini-4-carbon-vs-gpt-6-2-astra/. Accessed November 21, 2026.
Third-party data follows the license of each source. The full table is available as CSV and JSON.