Claude 3 Haiku
Claude / Anthropic
32scores
30benchmarks
$0.25 / $1.25 per 1M tokenscost in/out
Metadata
Claude Closed/API
Aliases: anthropic-claude-3-haiku, anthropic/claude-3-haiku, claude-3-haiku
| Benchmark | Category | Rank | Score | Sampled |
|---|---|---|---|---|
| Tau2-Bench Telecom | Agentic | 312 | 21.1% | 2026-05-11 |
| Terminal-Bench Hard | Agentic | 338 | 0.8% | 2026-05-11 |
| ToolSandbox | Agentic | 7 | 54.9 | 2026-05-27 |
| BigCodeBench | Coding | 57 | 39.40 | 2026-05-06 |
| ENAMEL | Coding | 9 | 0.39 | 2026-05-06 |
| EvalPlus | Coding | 24 | 68.85 | 2026-05-05 |
| LiveCodeBench | Coding | 28 | 20.20 | 2026-05-06 |
| Long Code Arena | Coding | 7 | 0.42 | 2026-05-06 |
| MBPP+ | Coding | 21 | 68.80 | 2026-05-05 |
| SciCode | Coding | 376 | 18.6% | 2026-05-11 |
| GSMA Open Telco Leaderboard | Domain | 56 | 45.84 | 2026-05-06 |
| BenchLM | General Knowledge | 102 | 24 | 2026-05-06 |
| MixEval Chat | General Knowledge | 31 | 42.80 | 2026-05-06 |
| AgentHarm | Generalization | 4 | 6.6% | 2026-05-27 |
| AgentHarm | Generalization | 6 | 11.1% | 2026-05-27 |
| AgentHarm | Generalization | 18 | 33.9% | 2026-05-27 |
| HELM AIR-Bench | Generalization | 18 | 0.827011 | 2026-05-28 |
| HELM Safety | Generalization | 42 | 0.877981 | 2026-05-28 |
| WildBench | Generalization | 35 | 7.0126953125 | 2026-05-27 |
| RubricEval | Instruction Following | 8 | 2.73 | 2026-05-06 |
| Artificial Analysis Intelligence Index | Intelligence | 387 | 12.26 | 2026-05-11 |
| HELM Lite | Intelligence | 65 | 0.294206 | 2026-05-28 |
| Humanity's Last Exam | Intelligence | 421 | 3.9% | 2026-05-11 |
| MathVista | Intelligence | 43 | 46.40 | 2026-05-06 |
| HindiGen v1 | Language | 21 | 55.32 | 2026-05-06 |
| BenchBench | Meta | 73 | 0.45 | 2026-05-06 |
| DROP | Reasoning | 17 | 0.78 | 2026-05-06 |
| GPQA Diamond | Reasoning | 407 | 37.4% | 2026-05-11 |
| ZebraLogic | Reasoning | 46 | 14.30 | 2026-05-06 |
| X-Risks Leaderboard | Safety | 9 | 13.06 | 2026-05-06 |
| CritPt | Science | 158 | 0% | 2026-05-11 |
| VNTL Leaderboard | Translation | 22 | 67.19 | 2026-05-06 |
No matching rows.