📡 LLM Pulse|🆕 New on leaderboard: Baidu: ERNIE-Light-DT-1.0 by Baidu — $1.13/M tokensLearn more →1/10

Measurement sources and verification status (17 fields)

Collected third-party figures are not independent tests. A source link and collection date do not establish reproducibility. Exact model version, dataset and test settings may be missing. No independent verification date is supplied.

speed: 80

latency: 0.32

contextWindow: 200000

inputPrice: 1

outputPrice: 5

benchmarkMmlu: 82

benchmarkHumanEval: 84.5

benchmarkMath: 71

benchmarkGpqa: 44

reasoningScore: 62

codingScore: 84.5

benchmarkSweBench: 38

aaIntelligenceIndex: 24

aaCostPerTask: 0.22

aaOutputSpeed: 99

aaLatency: 0.82

aaEndToEndTime: 5.85

Anthropic: Claude Haiku 4.5

anthropic🔒 Proprietary

Published Artificial Analysis leaderboard evidence.

LLM Stats Scores

62.0

Reasoning

84.5

Coding

🎯

44.0%GPQA Diamond — graduate-level science accuracy

Published benchmark evidence

Task scores are reported independently by each publisher; they are not combined into a hidden quality score.

View HLE source ↗

10.4%

Humanity's Last Exam

38.0%

SWE-bench

Artificial Analysis measurements

Published operational results. Cost per task, speed, and latency are kept separate from provider API prices.

Intelligence

AA Index

Cost / task

$0.22

AA measured

Output speed

tokens / second

TTFT

0.82s

AA measured

Response time

5.85s

end to end

Quality Score

72.0

Overall (0–100)

Speed

80 c/s

characters/second

Latency (TTFT)

0.32s

to first token

Input Price

per 1M tokens

Output Price

per 1M tokens

Context Window

200K

tokens

Release Date

2025-10-15

30-Day Rank History

📊 Benchmark Scores

Reported scores · % accuracy

What do these benchmarks test?

🎯

GPQA Diamond44.0%Below avg

Designed so that even PhD-level human experts average ~65–70%. A score above 80% signals near-expert scientific reasoning that goes far beyond memorisation.

🧠

MMLU82.0%Solid

The classic academic breadth benchmark. Top frontier models now score above 85%, so it best distinguishes mid-tier from strong generalist models.

💻

HumanEval84.5%Solid

Measures whether a model writes functionally correct code — not just plausible-looking syntax. Strong scores (>85%) indicate reliable day-to-day coding assistance.

∑

MATH Benchmark71.0%Solid

Tests multi-step mathematical reasoning under strict correctness. Even a 60% score requires strong chain-of-thought; top reasoning models now exceed 90%.