Anthropic
Claude Haiku 4.5
Announced Oct 15, 2025
Capable coding in a smaller model
What could it do?
Anthropic reported coding performance similar to Sonnet 4 at lower cost and latency.
What changed?
Brought a newer capability level to the Haiku tier.
Made frequent, smaller AI tasks more economical.
Where it fell short
Performance depends on the task; similar results on tests do not mean identical models.
Release facts were checked against the sources below. Performance claims belong to the developers; we have not independently tested these models.
Developer announcement or release logBenchmark results
A benchmark estimate, not a percentage or capability multiplier. Reasoning settings are not specified in this source table. Historical estimates can change in later snapshots.
Epoch AI methodology ↗Download the source snapshotChecked Oct 7, 2026 · CC BY 4.0Selected highest effort variant where available in the current table. Starred partial results are excluded. Claude fallback configurations can use other models when safeguards intervene.
Source: Artificial Analysis Intelligence Index ↗Download selected resultsChecked Oct 8, 2026One explicitly named variant per release. Scores come from the same Overall snapshot; preliminary entries and reported intervals are preserved. These are current ratings of earlier variants, not their launch day ratings.
Source: Text Arena ↗Download selected resultsChecked Oct 8, 2026Only mini-SWE-agent v2.0.0 submissions, one run per model. Reasoning effort varies and is labelled. These are published submissions, not a rerun by Leapscope. Latest included release: February 5, 2026; latest run: February 26, 2026.
Source: SWE bench Verified ↗Download selected resultsChecked Oct 8, 2026