📡 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: 160

latency: 0.2

contextWindow: 131072

inputPrice: 0.05

outputPrice: 0.33

benchmarkMmlu: 63.4

benchmarkHumanEval: 64

benchmarkMath: 42

benchmarkGpqa: 22

reasoningScore: 35

codingScore: 64

benchmarkSweBench: 15

Meta: Llama 3.2 3B Instruct

meta-llama🔒 Proprietary

Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it...

LLM Stats Scores

35.0

Reasoning

64.0

Coding

🎯

22.0%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 ↗

5.3%

Humanity's Last Exam

15.0%

SWE-bench

Quality Score

49.0

Overall (0–100)

Speed

160 c/s

characters/second

Latency (TTFT)

0.20s

to first token

Input Price

$0.05

per 1M tokens

Output Price

$0.33

per 1M tokens

Context Window

131K

tokens

Release Date

2024-09-25

30-Day Rank History

📊 Benchmark Scores

Reported scores · % accuracy

What do these benchmarks test?

🎯

GPQA Diamond22.0%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.

🧠

MMLU63.4%Below avg

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

💻

HumanEval64.0%Below avg

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

∑

MATH Benchmark42.0%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%.