📡 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 (12 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: 0.48

contextWindow: 32768

inputPrice: 0.36

outputPrice: 0.4

benchmarkMmlu: 86.1

benchmarkHumanEval: 86.4

benchmarkMath: 83.1

benchmarkGpqa: 49

reasoningScore: 66

codingScore: 86.4

benchmarkSweBench: 41

Qwen2.5 72B Instruct

qwen🔒 Proprietary

LLM Stats Scores

66.0

Reasoning

86.4

Coding

🎯

49.0%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 ↗

3.6%

Humanity's Last Exam

41.0%

SWE-bench

Quality Score

76.0

Overall (0–100)

Speed

80 c/s

characters/second

Latency (TTFT)

0.48s

to first token

Input Price

$0.36

per 1M tokens

Output Price

$0.4

per 1M tokens

Context Window

33K

tokens

Release Date

2024-09-19

30-Day Rank History

📊 Benchmark Scores

Reported scores · % accuracy

What do these benchmarks test?

🎯

GPQA Diamond49.0%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.

🧠

MMLU86.1%Solid

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

💻

HumanEval86.4%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 Benchmark83.1%Top tier

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