| Beam 1.0 | Claude Fable 5.2 | |
|---|---|---|
| CTO Index | 27.7 (#39) | 95.5 (#1) |
| Provider | Reflection AI | Anthropic |
| Weights | Open weights | Closed |
| Price $/M in / out | $0.30 / $1 | $8 / $40 |
| Context window | 1M | 1M |
| License | Apache 2.0 | Proprietary |
| Terminal-Bench 4.1 | 6.4% | 65.8% |
| WebDev Arena | 1437 | 1824 |
| ARC-AGI-2 | 23.5% | 93.8% |
| Text Arena | 1434 | 1503 |
| ARC-AGI-3 | — | 81.2% |
Snapshot 22 November 2026 · Scores copied from each benchmark's own leaderboard · How the CTO Index works
Frequently asked questions
Is Beam 1.0 better than Claude Fable 5.2?
Claude Fable 5.2 scores higher on the CTO Index (95.5 vs 27.7). Across the 4 benchmarks both have results, Beam 1.0 wins 0 and Claude Fable 5.2 wins 4 (Terminal-Bench, WebDev, ARC-AGI-2, Text Arena). Beam 1.0 costs 97% less per blended token.
Which is better for coding agents, Beam 1.0 or Claude Fable 5.2?
On Terminal-Bench 4.1, Claude Fable 5.2 scores 65.8% against Beam 1.0's 6.4%. Differences under about three points are within the error bars.
How do Beam 1.0 and Claude Fable 5.2 compare on price?
Beam 1.0 lists at $0.30 / $1 and Claude Fable 5.2 at $8 / $40 per million input / output tokens.