Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Anthropic · Proprietary · rank 191 · bench-align-v5
Claude Haiku 4.5 is a general-purpose language model. It accepts text, images and PDFs and produces text.
Claude Haiku 4.5 by Anthropic scores 33.8 overall, costs $1.00 per 1M input tokens and $5.00 per 1M output tokens, and carries a context window of 200K tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Every published price component, including cache reads and writes.
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
Lifecycle events the source associates with this model, plus what this profile does not claim.
That is not evidence the model has no lifecycle plan — only that this source published none.
Each axis shows the ranking source’s score and, beside it, BenchLM’s separate verified score for that axis. The two are published side by side and are not interchangeable.
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: ObservedUnavailable
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | Unavailable | Unavailable | Unavailable | |
| Throughput | Unavailable | Unavailable | Unavailable |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
Plans from the model's maker, and the API providers that list this exact model today. Links leave ModelChoice; each provider sets its own prices and limits.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| GPQA Diamond (Vals) GPQA Diamond, Vals AI run | 72.2 | #74 | Graduate-level science questions | Expert reasoning | Vals AI GPQA Diamond, Vals AI run leaderboard |
| MMLU-Pro (Vals) MMLU-Pro, Vals AI run | 78.7 | #77 | Academic multiple-choice questions | Broad academic knowledge | Vals AI MMLU-Pro, Vals AI run leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| SWE-bench Verified Software Engineering Benchmark Verified | 73.3 | #58 | 500 verified issues | Professional software engineering | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? |
| VulcanBench v3 VulcanBench v3 | 76.2 | #13 | 23 post-cutoff repository tasks in the v3 report | Professional multi-file software engineering | VulcanBench |
| LiveCodeBench (Vals) LiveCodeBench, Vals AI run | 41.2 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3) FrontierMath v2 Tiers 1-3 | 5.903 | #84 | 285 private advanced mathematics problems | From olympiad-plus to early research mathematics | FrontierMath v2 benchmark hub |
| FrontierMath v2 (Tier 4) FrontierMath v2 Tier 4 | 2.083 | #70 | 41 private extreme-difficulty mathematics problems | Research-level mathematics requiring hours or days of expert work | FrontierMath Tier 4 v2 leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| JobBench JobBench | 16.0 | #34 | 130 tasks across 35 occupations | Professional multi-source workflows | JobBench: Aligning Agent Work With Human Will |
| Terminal-Bench 2.1 (Vals) Terminal-Bench 2.1, Vals AI run | 43.8 | #61 | Difficult terminal tasks | Frontier agentic | Vals AI Terminal-Bench 2.1, Vals AI run leaderboard |
| MazeBench MazeBench: visual spatial reasoning in a 3D open world | 0 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Design Arena Website Design Arena Website Elo | 1110 | #98 | Website generation comparisons | Design and website generation | OpenRouter Grok 4.3 benchmarks |
| ParseBench-1 ParseBench-1: document parsing for AI agents (LlamaIndex leaderboard, first release) | 49.34 | #35 | 2,078 human-verified enterprise pages from insurance, finance and government documents | No method is strong on all five dimensions; charts and visual grounding separate the field | ParseBench: A Document Parsing Benchmark for AI Agents |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Chess Puzzles Chess Puzzles (Epoch AI) | 8.0 | #82 | 100 programmatically generated chess positions with a single best move | Novel chess positions, single best move | Chess Puzzles (Epoch AI) |
| FailureSensorIQ FailureSensorIQ | 61.8 | #14 | Multiple-choice questions on sensors and failure modes for 10 industrial assets | Industrial maintenance and reliability knowledge | BenchLM |
| BinaryAudit BinaryAudit (backdoor detection in compiled binaries) | 10.5 |
| #91 |
| Competitive programming problems (easy, medium, hard) |
| Frontier coding |
| Vals AI LiveCodeBench, Vals AI run leaderboard |
| SWE-bench (Vals) SWE-bench, Vals AI run | 66.6 | #65 | Real repository issues by human time bucket | Frontier coding agents | Vals AI SWE-bench, Vals AI run leaderboard |
| #29 |
| 100 hidden gems across 256 rooms in one open world |
| Long-horizon spatial planning, with single puzzles that can take over one hundred moves |
| MazeBench Results |
| OSWorld 2.1 Offline OSWorld 2.1 offline subset (82 tasks) | 15.7 | #11 | 82 offline computer-use workflows | Long-horizon professional workflows | OSWorld 2.1 release (xlang-ai/OSWorld-V2, tag osworld-v2.1) |
| #16 |
| 19 backdoor and timebomb detection tasks on stripped binaries, three runs each (57 runs); 6 tooling and 8 clean-binary tasks reported alongside |
| Reverse engineering without source code or symbols |
| We hid backdoors in ~40 MB binaries and asked AI + Ghidra to find them (Quesma) |
| OTelBench OTelBench (OpenTelemetry instrumentation tasks) | 5.8 | #11 | 23 OpenTelemetry instrumentation tasks in 11 languages, three attempts each (69 runs) | Distributed tracing and context propagation across services | Benchmarking OpenTelemetry: can AI trace your failed login? (Quesma) |
| CompileBench CompileBench (building real projects from source, pass@1) | 68.9 | #7 | 15 build tasks on cowsay, jq, GNU coreutils and curl, three attempts each (45 attempts) | Dependency resolution, legacy toolchains, static linking and cross-compilation | CompileBench: can AI compile 22-year-old code? (Quesma) |
| NYT Connections NYT Connections LLM Benchmark (extended version) | 14.3 | #87 | 940 NYT Connections puzzles, each with up to four added trick words | Wordplay, word knowledge and decoys that fit almost-but-not-quite | NYT Connections LLM Benchmark (Lech Mazur, GitHub) |
| SkillsBench official leaderboard (BenchFlow) SkillsBench v1.1 official leaderboard (BenchFlow) | 30.1 | #20 | 87 tasks in 8 domains (SkillsBench v1.1) | Multi-step agent tasks | SkillsBench leaderboard |