| Gemini 4 Carbon | Claude Fable 5.2 | |
|---|---|---|
| CTO Index | 92.4 (#3) | 95.5 (#1) |
| Provider | Anthropic | |
| Weights | Closed | Closed |
| Price $/M in / out | $2 / $10 | $8 / $40 |
| Context window | Not listed | 1M |
| License | Proprietary | Proprietary |
| Terminal-Bench 4.1 | 64.5% | 65.8% |
| WebDev Arena | 1769 | 1824 |
| ARC-AGI-2 | 94.2% | 93.8% |
| Text Arena | 1527 | 1503 |
| ARC-AGI-3 | 65.1% | 81.2% |
Snapshot 22 November 2026 · Scores copied from each benchmark's own leaderboard · How the CTO Index works
Frequently asked questions
Is Gemini 4 Carbon better than Claude Fable 5.2?
Claude Fable 5.2 scores higher on the CTO Index (95.5 vs 92.4). Across the 5 benchmarks both have results, Gemini 4 Carbon wins 2 (ARC-AGI-2, Text Arena) and Claude Fable 5.2 wins 3 (Terminal-Bench, WebDev, ARC-AGI-3). Gemini 4 Carbon costs 75% less per blended token.
Which is better for coding agents, Gemini 4 Carbon or Claude Fable 5.2?
On Terminal-Bench 4.1, Claude Fable 5.2 scores 65.8% against Gemini 4 Carbon's 64.5%. Differences under about three points are within the error bars.
How do Gemini 4 Carbon and Claude Fable 5.2 compare on price?
Gemini 4 Carbon lists at $2 / $10 and Claude Fable 5.2 at $8 / $40 per million input / output tokens.