Qwen2 72B Instruct — Benchmarks

Benchmark scores for Qwen2 72B Instruct aggregated from public leaderboards (7 boards), with how it ranks among open models and how it compares with its nearest neighbours. See hardware requirements for what you need to run it.

llmrun Composite Score →

77.5

90% range 71.1–83.8 · from 3 boards

CodingNot enough data yet
AgentsNot enough data yet
MathNot enough data yet
ScienceNot enough data yet
ReasoningNot enough data yet

Qwen2 72B Instruct reasoning benchmarks

Multi-step logic and problem-solving — can the model think through a hard problem rather than recall an answer.

GPQA Diamond
GPQA Diamond is a set of extremely hard, graduate-level science questions (physics, chemistry, biology) written by domain experts and filtered so that skilled non-experts with web access still fail. It measures genuine reasoning rather than memorization.
40.8%
40.8%
median 55.7%open leader 93.2%
Overall best: Gemini 4 Argon (high) 96.4%
29th percentile · #71 of 100 open · #262 of 309 all
#71 of 100
#262 of 309
Independent
Epoch AI · Jun 2024

Qwen2 72B Instruct coding benchmarks

Writing, editing, and fixing real code — measured by whether the resulting program actually runs and passes tests.

WeirdML
WeirdML asks models to write machine-learning code for small, deliberately unusual datasets, then runs the code and scores the accuracy it actually achieves. Built by Håvard Tveit Ihle, it rewards practical problem-solving over recognizing a familiar task.
11.3%
11.3%
median 39.5%open leader 82.6%
Overall best: GPT 6 Astra Promax 93.6%
7th percentile · #44 of 47 open · #155 of 162 all
#44 of 47
#155 of 162
Independent
Epoch AI · Jun 2024

Other results (1)leaderboard-only, not part of the llmrun Score

Leaderboard-only, not part of the llmrun Score

METR Time Horizons
METR Time Horizons, from the AI-evaluation nonprofit METR, runs models on a suite of software-engineering and research tasks whose length is measured by how long they take skilled humans, from seconds to many hours. METR summarises each model as a time horizon: the human task length at which the model succeeds half the time. llmrun shows METR's average task-success score (0–1) as published by Epoch AI; higher means the model completes more of the suite, including longer tasks.
29.9%
29.9%
median 49.6%open leader 59.2%
Overall best: Claude Mythos Preview Early 85.2%
13th percentile · #8 of 9 open · #39 of 46 all
#8 of 9
#39 of 46
Lab-reported
Epoch AI · Jun 2024

Qwen2 72B Instruct math benchmarks

Symbolic and quantitative problem-solving, from competition math to multi-step calculation.

MATH Level 5
MATH Level 5 covers the hardest tier of competition-mathematics problems, testing multi-step symbolic and quantitative reasoning.
39.1%
39.1%
median 41.6%open leader 96.6%
Overall best: GPT 5 (Aug 07, 2025, high) 98.1%
45th percentile · #22 of 39 open · #75 of 107 all
#22 of 39
#75 of 107
Independent
Epoch AI · Jun 2024

Qwen2 72B Instruct knowledge benchmarks

Breadth and depth of factual knowledge across many subjects, paired with the reasoning to apply it.

MMLU-Pro
MMLU-Pro is a harder, cleaned-up successor to MMLU with ten answer choices and more reasoning-heavy questions across 14 subjects, measuring broad knowledge and reasoning together.
64.4%
64.4%
median 52.0%open leader 88.0%
Overall best: Gemini-3.1-Pro 91.2%
67th percentile · #49 of 146 open · #120 of 257 all
#49 of 146
#120 of 257
Independent
TIGER-Lab

Other results (1)leaderboard-only, not part of the llmrun Score

Leaderboard-only, not part of the llmrun Score

MMLU
MMLU (Massive Multitask Language Understanding) spans 57 subjects from history to law to medicine as multiple-choice questions. It is the long-standing default for broad knowledge and remains the most widely-reported general benchmark.
82.4%
82.4%
median 62.5%open leader 87.2%
Overall best: GPT 4o (Nov 20, 2024) 88.1%
92nd percentile · #8 of 85 open · #19 of 136 all
#8 of 85
#19 of 136
Lab-reported
Epoch AI · Jun 2024

Qwen2 72B Instruct instruction following benchmarks

How precisely a model obeys explicit, verifiable instructions — format, length, and content constraints — and, more broadly, how much people prefer its responses head-to-head in blind human voting.

LMArena Text
Arena (arena.ai, formerly LMArena / Chatbot Arena) ranks models by millions of blind, head-to-head human votes on real prompts. The Arena Score shown here is a Bradley-Terry rating computed with Arena's default style control, so higher means people preferred that model's answers more often — it's a relative rating, not a percentage, so compare it only to other models on this board.
1261.9
1261.9
median 1321.4open leader 1488.2
Overall best: Gemini 4 Carbon (High) 1527.5
33rd percentile · #127 of 188 open · #308 of 408 all
#127 of 188
#308 of 408
Arena
Arena · Oct 2026

Score vs VRAM in the ≤64 GB tier

Qwen2 72B Instruct
Other models
Best Score for its VRAM
x: VRAM (GB, log scale) · y: llmrun Score

#40 of 51 that fit ≤64 GB

Each dot is a model at the best quant that fits ≤64 GB. The line steps up to the best Score available at each amount of VRAM; a dot below it is beaten by a model that needs no more memory.

Full ≤64 GB ranking →

Qwen2 72B Instruct vs its neighbours

Same boards for the nearest open models per llmrun Score. The highest score in each row is marked.

Board	Qwen2 72B Instructthis model	Qwen3.8 27Bbest in classQ4_K_M · 18 GB	Llama 3.3 70B Instructnearest by ScoreQ4_K_M · 51 GB	C4ai Command A 03 2025nearest by ScoreQ4_K_M · 79 GB	Llama 3.1 70B Instructnearest by ScoreQ4_K_M · 51 GB
llmrun Score	77.5	122.0	80.0	79.8	77.9
GPQA Diamond	40.8%	—	47.4%	—	44.2%
WeirdML	11.3%	—	14.4%	—	9.0%
MATH Level 5	39.1%	—	41.6%	—	36.7%
MMLU-Pro	64.4%	—	65.9%	—	62.8%
LMArena Text	1261.9	1435.6	1318.0	1353.7	1293.4
Compare head to head		Compare	Compare	Compare	Compare

Scores aggregated from public benchmark sources (Epoch AI: 5 boards, Arena: 1 board, TIGER-Lab: 1 board) — each linked from the benchmark pages. llmrun does not run these benchmarks. Percentiles and ranks are among open models in the llmrun catalogue; the median and leader marks refer to the same field.