📡 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 (14 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: 0.25

outputPrice: 1.25

benchmarkMmlu: 75.2

benchmarkHumanEval: 75.9

benchmarkMath: 38.9

benchmarkHellaSwag: 84.3

benchmarkGpqa: 33.3

benchmarkMmluPro: 45

reasoningScore: 62

codingScore: 84.5

benchmarkSweBench: 38

Anthropic: Claude 3 Haiku

anthropic🔒 Proprietary

LLM Stats Scores

62.0

Reasoning

84.5

Coding

🎯

33.3%GPQA Diamond — graduate-level science accuracy

45.0%MMLU-Pro

Published benchmark evidence

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

View HLE source ↗

4.1%

Humanity's Last Exam

38.0%

SWE-bench

BenchLM performance measurements

Published output speed and first-token timing, kept separate from provider pricing and other benchmark sources.

View source ↗

Output speed

138

tokens / second

TTFT

1.16s

BenchLM measured

Collected 9/25/2026

Quality Score

72.0

Overall (0–100)

Speed

80 c/s

characters/second

Latency (TTFT)

0.32s

to first token

Input Price

$0.25

per 1M tokens

Output Price

$1.25

per 1M tokens

Context Window

200K

tokens

Release Date

2024-03-13

30-Day Rank History

📊 Benchmark Scores

Reported scores · % accuracy

What do these benchmarks test?

🎯

GPQA Diamond33.3%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.

📚

MMLU-Pro45.0%Below avg

The harder successor to MMLU — more options per question eliminate single-answer guessing. Tests depth of professional knowledge across 14 disciplines.

🧠

MMLU75.2%Solid

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

💻

HumanEval75.9%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 Benchmark38.9%Below avg

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

🌐

HellaSwag84.3%Below avg

Tests grounded, everyday common sense rather than memorised facts. Most frontier models now score 90%+, so it best separates small/mid models from large ones.