# Claude Haiku 4.5 benchmark scores

Anthropic · model · https://benchgap.net/model/claude-haiku-4-5 · data retrieved 2026-11-22

As of 2026-11-22, Claude Haiku 4.5 (Anthropic) has measured scores on 18 benchmarks and estimated scores on 96 more.

| Benchmark | Score | Source |
|---|---|---|
| [AA-GPQA Diamond](https://benchgap.net/b/aa-gpqa-diamond/current.md) | 74.4% | estimated ± 3.6 pp, high confidence |
| [AA-HLE](https://benchgap.net/b/aa-hle/current.md) | 9.3% | estimated ± 6.2 pp, medium confidence |
| [AA-SciCode](https://benchgap.net/b/aa-sci-code/current.md) | 45.3% | estimated ± 3.8 pp, high confidence |
| [Artificial Analysis Intelligence Index](https://benchgap.net/b/artificialanalysis/current.md) | 12.6% | estimated ± 5.8 pp, medium confidence |
| [CritPt](https://benchgap.net/b/critpt/current.md) | 0.9% | estimated ± 5.3 pp, medium confidence |
| [GDPval-AA](https://benchgap.net/b/gdpvalaanormalized/current.md) | 45.5% | estimated ± 1.7 pp, low confidence |
| [GPQA Diamond](https://benchgap.net/b/gpqa-diamond/current.md) | 80.1% | estimated ± 1.8 pp, medium confidence |
| [HLE](https://benchgap.net/b/hle/current.md) | 25.9% | estimated ± 9.2 pp, low confidence |
| [AA-Omniscience Accuracy](https://benchgap.net/b/omniscienceaccuracy/current.md) | 22.2% | estimated ± 6.5 pp, medium confidence |
| [SciCode](https://benchgap.net/b/scicode/current.md) | 44.5% | estimated ± 3.6 pp, high confidence |
| [Terminal-Bench 2.1](https://benchgap.net/b/terminal-bench%23terminal-bench-2-1/current.md) | 45.5% | estimated ± 6.7 pp, low confidence |
| [BrowseComp](https://benchgap.net/b/browsecomp/current.md) | 66.5% | estimated ± 6.2 pp, medium confidence |
| [HLE w/ tools](https://benchgap.net/b/hlewithtools/current.md) | 53.7% | estimated ± 7.5 pp, low confidence |
| [AA Coding Index](https://benchgap.net/b/aa-coding-index/current.md) | 46.7% | estimated ± 5.3 pp, medium confidence |
| [AA-MMMU-Pro](https://benchgap.net/b/aa-mmmu-pro/current.md) | 64.0% | estimated ± 6.1 pp, medium confidence |
| [APEX-Agents-AA](https://benchgap.net/b/apexagentsaa/current.md) | 19.9% | estimated ± 3.6 pp, medium confidence |
| [DeepSearchQA](https://benchgap.net/b/deepsearchqa/current.md) | 75.2% | estimated ± 8.2 pp, low confidence |
| [SWE-bench Verified](https://benchgap.net/b/swe-bench-verified/current.md) | 73.3% | measured |
| [CyberGym](https://benchgap.net/b/cybergym/current.md) | 37.9% | estimated ± 11.4 pp, low confidence |
| [JobBench](https://benchgap.net/b/jobbench/current.md) | 16.0% | measured |
| [SWE-bench Pro](https://benchgap.net/b/swe-bench-pro/current.md) | 51.1% | estimated ± 4.5 pp, high confidence |
| [WideResearch](https://benchgap.net/b/wideresearch/current.md) | 73.8% | estimated ± 1.2 pp, medium confidence |
| [MMLU-Pro](https://benchgap.net/b/mmlu-pro/current.md) | 81.1% | estimated ± 4.8 pp, high confidence |
| [FailureSensorIQ](https://benchgap.net/b/failuresensoriq/current.md) | 61.8% | measured |
| [GPQA](https://benchgap.net/b/gpqa/current.md) | 75.1% | estimated ± 4.5 pp, high confidence |
| [MMLU](https://benchgap.net/b/mmlu/current.md) | 88.4% | estimated ± 1.2 pp, high confidence |
| [Vals GPQA Diamond](https://benchgap.net/b/vals-gpqa-diamond/current.md) | 72.2% | measured |
| [Vals LiveCodeBench](https://benchgap.net/b/valslivecodebench/current.md) | 41.2% | measured |
| [Vals MMLU-Pro](https://benchgap.net/b/vals-mmlu-pro/current.md) | 78.7% | measured |
| [Chess Puzzles](https://benchgap.net/b/chesspuzzles/current.md) | 8.0% | measured |
| [AA MMLU-Pro](https://benchgap.net/b/aa-mmlu-pro/current.md) | 83.5% | estimated ± 1.1 pp, medium confidence |
| [Gert Labs](https://benchgap.net/b/gertlabs/current.md) | 33.6% | estimated ± 7.7 pp, medium confidence |
| [AA-AnalystAgent](https://benchgap.net/b/aa-analyst-agent/current.md) | 31.4% | estimated ± 2.7 pp, low confidence |
| [AA EnterpriseOps-Gym](https://benchgap.net/b/aa-enterprise-ops-gym/current.md) | 40.4% | estimated ± 1.2 pp, low confidence |
| [AA Terminal-Bench 2.1](https://benchgap.net/b/aa-terminal-bench21/current.md) | 79.4% | estimated ± 2.5 pp, medium confidence |
| [ApprenticeBench](https://benchgap.net/b/apprenticebench/current.md) | 0.1% | estimated ± 6.0 pp, low confidence |
| [ARC-AGI-1](https://benchgap.net/b/arcagi1/current.md) | 49.3% | estimated ± 6.5 pp, low confidence |
| [ARC-AGI-2](https://benchgap.net/b/arc-agi-2/current.md) | 0.0% | estimated ± 13.8 pp, low confidence |
| [CursorBench 3.1](https://benchgap.net/b/cursor-bench31/current.md) | 40.4% | estimated ± 5.8 pp, low confidence |
| [CursorBench 3.2](https://benchgap.net/b/cursor-bench32/current.md) | 61.5% | estimated ± 3.3 pp, medium confidence |
| [EBR-bench](https://benchgap.net/b/ebrbench/current.md) | 0.0% | estimated ± 9.1 pp, low confidence |
| [FrontierCode 1.1 Main](https://benchgap.net/b/frontiercode/current.md) | 15.7% | estimated ± 4.2 pp, low confidence |
| [FrontierSWE v2](https://benchgap.net/b/frontierswev2/current.md) | 5.5% | estimated ± 9.5 pp, low confidence |
| [Furniture Assembly](https://benchgap.net/b/furnitureassembly/current.md) | 24.7% | estimated ± 14.1 pp, low confidence |
| [MaxxBench](https://benchgap.net/b/maxxbench/current.md) | 32.8% | estimated ± 10.3 pp, low confidence |
| [MazeBench](https://benchgap.net/b/mazebench/current.md) | 0.0% | measured |
| [Mystery Game Puzzles](https://benchgap.net/b/mysterygamepuzzles/current.md) | 8.7% | estimated ± 2.2 pp, low confidence |
| [NYT Connections](https://benchgap.net/b/nyt-connections/current.md) | 14.3% | measured |
| [OfficeQA Pro](https://benchgap.net/b/officeqapro/current.md) | 47.1% | estimated ± 7.0 pp, low confidence |
| [OSWorld-Verified](https://benchgap.net/b/osworld-verified/current.md) | 60.8% | estimated ± 8.4 pp, low confidence |
| [ParseBench-1](https://benchgap.net/b/parse-bench-llama-index/current.md) | 49.3% | measured |
| [Terminal-Bench 4.1](https://benchgap.net/b/terminal-bench41/current.md) | 6.1% | estimated ± 14.9 pp, low confidence |
| [Terminal-Bench Hard](https://benchgap.net/b/terminal-bench-hard/current.md) | 44.0% | estimated ± 3.9 pp, medium confidence |
| [Vals SWE-bench](https://benchgap.net/b/valsswebench/current.md) | 66.6% | measured |
| [Terminal-Bench 2.1 (Vals AI)](https://benchgap.net/b/vals-terminal-bench21/current.md) | 43.8% | measured |
| [VulcanBench v3](https://benchgap.net/b/vulcanbench%23vulcanbench-v3/current.md) | 76.2% | measured |
| [CWE-Bench-AA](https://benchgap.net/b/aa-cwe-bench/current.md) | 44.0% | estimated ± 10.6 pp, low confidence |
| [DeepsecBench-AA](https://benchgap.net/b/aa-deepsec-bench/current.md) | 21.9% | estimated ± 10.3 pp, low confidence |
| [Bug Hunt Bench](https://benchgap.net/b/bug-hunt-bench/current.md) | 19.0% | estimated ± 8.6 pp, low confidence |
| [CursorBench 4.0](https://benchgap.net/b/cursor-bench40/current.md) | 30.8% | estimated ± 3.4 pp, medium confidence |
| [CWE-bench v1](https://benchgap.net/b/cwe-bench%23cwe-bench-v1/current.md) | 57.4% | estimated ± 5.1 pp, medium confidence |
| [DeepSWE](https://benchgap.net/b/deepswe/current.md) | 56.3% | estimated ± 7.2 pp, low confidence |
| [DRACO](https://benchgap.net/b/draco/current.md) | 86.9% | estimated ± 0.7 pp, low confidence |
| [FrontierPhysics](https://benchgap.net/b/frontierphysics/current.md) | 0.0% | estimated ± 6.1 pp, low confidence |
| [HLE w/o tools](https://benchgap.net/b/hlenotools/current.md) | 24.6% | estimated ± 4.9 pp, high confidence |
| [OSWorld 2.1](https://benchgap.net/b/osworld-2-1/current.md) | 66.4% | estimated ± 6.1 pp, medium confidence |
| [OSWorld 2.1 Offline](https://benchgap.net/b/osworld-2-1-offline/current.md) | 15.7% | measured |
| [PostTrainBench v1.1](https://benchgap.net/b/posttrainbench%23posttrainbench-v1-1/current.md) | 12.5% | estimated ± 3.7 pp, medium confidence |
| [SpireClimb](https://benchgap.net/b/spireclimb/current.md) | 0.0% | estimated ± 10.3 pp, low confidence |
| [StealthMark](https://benchgap.net/b/stealthmark/current.md) | 30.2% | estimated ± 9.7 pp, low confidence |
| [SWE-bench Multimodal](https://benchgap.net/b/swe-bench-multimodal/current.md) | 0.0% | estimated ± 5.6 pp, low confidence |
| [Toolathlon-Verified](https://benchgap.net/b/toolathlonverified/current.md) | 73.4% | estimated ± 2.6 pp, high confidence |
| [AA AutomationBench](https://benchgap.net/b/aa-automation-bench/current.md) | 52.8% | estimated ± 12.8 pp, low confidence |
| [GDP.pdf](https://benchgap.net/b/aa-gdp-pdf/current.md) | 16.4% | estimated ± 4.1 pp, low confidence |
| [AA Terminal-Bench 4.0](https://benchgap.net/b/aa-terminal-bench4/current.md) | 0.0% | estimated ± 13.7 pp, low confidence |
| [CWE-bench](https://benchgap.net/b/cwe-bench-v2/current.md) | 53.9% | estimated ± 7.9 pp, low confidence |
| [MKA](https://benchgap.net/b/mka/current.md) | 37.3% | estimated ± 3.4 pp, medium confidence |
| [SkillsBench official leaderboard (BenchFlow)](https://benchgap.net/b/skillsbench/current.md) | 30.1% | measured |
| [BinaryAudit](https://benchgap.net/b/binaryaudit/current.md) | 10.5% | measured |
| [CompileBench](https://benchgap.net/b/compilebench/current.md) | 68.9% | measured |
| [OTelBench](https://benchgap.net/b/otelbench/current.md) | 5.8% | measured |
| [Vibe Code Bench](https://benchgap.net/b/vibecodebench/current.md) | 0.6% | estimated ± 8.4 pp, medium confidence |
| [FACTS Multimodal](https://benchgap.net/b/facts-multimodal/current.md) | 31.8% | estimated ± 5.6 pp, low confidence |
| [Hermes Index](https://benchgap.net/b/hermesindex/current.md) | 37.2% | estimated ± 3.1 pp, high confidence |
| [ParseBench-2](https://benchgap.net/b/parsebench/current.md) | 59.8% | estimated ± 4.3 pp, high confidence |
| [CharXiv](https://benchgap.net/b/charxiv/current.md) | 79.8% | estimated ± 5.6 pp, low confidence |
| [ExploitBench](https://benchgap.net/b/exploitbench/current.md) | 7.6% | estimated ± 13.4 pp, low confidence |
| [ExploitGym](https://benchgap.net/b/exploitgym/current.md) | 11.9% | estimated ± 6.3 pp, medium confidence |
| [Claw-Eval](https://benchgap.net/b/claw-eval/current.md) | 54.1% | estimated ± 4.5 pp, medium confidence |
| [MCP Atlas](https://benchgap.net/b/mcpatlas/current.md) | 38.1% | estimated ± 8.6 pp, medium confidence |
| [MMMU-Pro](https://benchgap.net/b/mmmu-pro/current.md) | 71.0% | estimated ± 6.9 pp, medium confidence |
| [NL2Repo](https://benchgap.net/b/nl2repo/current.md) | 38.9% | estimated ± 4.7 pp, medium confidence |
| [PostTrainBench v1](https://benchgap.net/b/posttrainbench%23posttrainbench-v1/current.md) | 7.8% | estimated ± 3.6 pp, medium confidence |
| [ScreenSpot Pro](https://benchgap.net/b/screenspot-pro/current.md) | 72.6% | estimated ± 10.1 pp, low confidence |
| [Terminal-Bench 2.0](https://benchgap.net/b/terminal-bench%23terminal-bench-2-0/current.md) | 53.4% | estimated ± 5.8 pp, medium confidence |
| [Toolathlon](https://benchgap.net/b/toolathlon/current.md) | 29.7% | estimated ± 7.4 pp, low confidence |
| [AA LiveCodeBench](https://benchgap.net/b/aa-live-code-bench/current.md) | 85.4% | estimated ± 2.6 pp, high confidence |
| [React Native Evals](https://benchgap.net/b/reactnativeevals/current.md) | 76.6% | estimated ± 2.7 pp, medium confidence |
| [ResearchClawBench](https://benchgap.net/b/researchclawbench/current.md) | 14.9% | estimated ± 2.6 pp, low confidence |
| [SWE-Rebench](https://benchgap.net/b/swe-rebench/current.md) | 57.0% | estimated ± 3.6 pp, medium confidence |
| [PostTrainBench](https://benchgap.net/b/post-train-bench12/current.md) | 10.3% | estimated ± 3.7 pp, medium confidence |
| [DAYJOB](https://benchgap.net/b/dayjob/current.md) | 0.4% | estimated ± 6.3 pp, low confidence |
| [DAYJOB Healthcare](https://benchgap.net/b/dayjob-healthcare/current.md) | 0.1% | estimated ± 6.1 pp, low confidence |
| [Finance Agent v2](https://benchgap.net/b/financeagentv2/current.md) | 56.1% | estimated ± 3.8 pp, medium confidence |
| [ScarfBench](https://benchgap.net/b/scarfbench/current.md) | 38.2% | estimated ± 12.3 pp, low confidence |
| [BlenderWorld (text-only)](https://benchgap.net/b/blenderworld-text-only/current.md) | 70.6% | estimated ± 4.0 pp, low confidence |
| [Cookie-Bench](https://benchgap.net/b/cookie-bench/current.md) | 73.4% | estimated ± 2.5 pp, low confidence |
| [LiveCodeBench](https://benchgap.net/b/livecodebench/current.md) | 75.1% | estimated ± 14.6 pp, low confidence |
| [Agents' Last Exam](https://benchgap.net/b/agentslastexam/current.md) | 35.4% | estimated ± 11.0 pp, low confidence |
| [OpenHarmony Bench v1.0](https://benchgap.net/b/openharmonybench%23openharmonybench-v1-0/current.md) | 50.1% | estimated ± 3.2 pp, medium confidence |
| [SEC-Bench Pro](https://benchgap.net/b/secbenchpro/current.md) | 13.4% | estimated ± 12.3 pp, low confidence |
| [LifeVal](https://benchgap.net/b/lifeval/current.md) | 75.2% | estimated ± 3.8 pp, medium confidence |
| [MMMU-Pro w/ Python](https://benchgap.net/b/mmmupropython/current.md) | 74.2% | estimated ± 2.7 pp, medium confidence |
| [ExploitBench Internal Port](https://benchgap.net/b/exploitbenchinternalport/current.md) | 0.0% | estimated ± 1.4 pp, low confidence |
