Claude Haiku 4.5
Anthropic · elicited April 2026 · average interval-width rank #8.8 of 29 across the canonical panel · $1.20 total elicitation cost.
Canonical panel profile
One row per canonical quantity, each on its own scale: this model's pooled center and 90 percent interval, with the 29-model panel median as a gray tick.
Generation harness
From the paper's per-model disclosure (Appendix A16). The prompt and repeated-run design are identical across models; the output mechanism follows each provider's API surface.
- Provider path
- LiteLLM
- Output mechanism
- forced function call
- Completion budget
- 1200
- Sampling
- temperature 1.0
- Reasoning
- off (provider default)
- API identifier
- claude-haiku-4-5-20251001 (dated snapshot)
195 successful runs · elicited April and July 2026 · v4 prompts · 15 runs per model-quantity cell. Code · Raw responses · Paper (PDF)
Code and dataElicited April and July 2026 · 29 models · v4 prompts