Qwen2.5 7B Instruct — Benchmarks

Benchmark scores for Qwen2.5 7B Instruct aggregated from public leaderboards (6 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 →

65.1

90% range 55.0–75.3 · from 5 boards

CodingNot enough data yet
AgentsNot enough data yet
MathNot enough data yet
ScienceNot enough data yet
Reasoning65.290% range 51–80

Qwen2.5 7B 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.
35.5%
35.5%
median 54.0%open leader 93.1%
Overall best: Gemini 4 Argon (high) 96.4%
24th percentile · #75 of 98 open · #266 of 300 all
#75 of 98
#266 of 300
Independent
Epoch AI · Sep 2024
BALROG
BALROG (Benchmarking Agentic LLM and VLM Reasoning On Games) tests long-horizon agentic reasoning across six game environments — BabyAI, Crafter, TextWorld, Baba Is AI, MiniHack and NetHack — scoring the average progress a model makes toward completing each game. It was built by UCL's DARK lab and collaborators.
7.8%
7.8%
median 16.2%open leader 34.9%
Overall best: GPT 6 Astra Max 68.3%
14th percentile · #13 of 15 open · #38 of 40 all
#13 of 15
#38 of 40
Independent
Epoch AI · Sep 2024
DTBench
DTBench is a set of handcrafted, expert-validated multiple-choice questions on decision theory — Newcomb-like problems and related puzzles — built by the Conceptual Reasoning Index project with Anthropic. It tests whether a model can follow an abstract argument rather than recall a fact.
47.7%
47.7%
median 67.7%open leader 93.9%
Overall best: Claude Opus 5.5 Max 98.9%
6th percentile · #64 of 68 open · #206 of 214 all
#64 of 68
#206 of 214
Lab-reported
Epoch AI · Sep 2024
LMCA
LMCA (Language Model Conceptual Argumentation) scores models on expert-rated arguments across philosophy, decision theory, and AI risk — domains where there is no empirical feedback loop to check an answer against. It is part of the Conceptual Reasoning Index, built with Anthropic by researchers Emery Cooper and Caspar Oesterheld.
6.4%
6.4%
median 24.7%open leader 55.5%
Overall best: Claude Opus 5.5 Max 68.2%
8th percentile · #50 of 54 open · #170 of 176 all
#50 of 54
#170 of 176
Lab-reported
Epoch AI · Sep 2024

Qwen2.5 7B Instruct math benchmarks

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

AIME 2024/2025
AIME (American Invitational Mathematics Examination) is a prestigious high-school competition of hard, integer-answer problems. It's a widely-cited yardstick for multi-step mathematical reasoning, here on the 2024–2025 papers.
2.5%
2.5%
median 52.8%open leader 98.6%
Overall best: Claude Fable 5 (high) 100.0%
14th percentile · #75 of 86 open · #265 of 283 all
#75 of 86
#265 of 283
Independent
Epoch AI · Sep 2024

Qwen2.5 7B Instruct knowledge benchmarks

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

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.
72.9%
72.9%
median 62.5%open leader 87.2%
Overall best: GPT 4o (Nov 20, 2024) 88.1%
73rd percentile · #24 of 85 open · #53 of 136 all
#24 of 85
#53 of 136
Lab-reported
Epoch AI · Sep 2024

Score vs VRAM in the 24 GB tier

Qwen2.5 7B Instruct
Other models
Best Score for its VRAM
x: VRAM (GB) · y: llmrun Score

#32 of 35 that fit 24 GB

Each dot is a model at the best quant that fits a 24 GB GPU. The line joins the models with the best Score for their VRAM; a dot below it is beaten by a model that needs no more memory.

Full 24 GB ranking →

Qwen2.5 7B Instruct vs its neighbours

Same boards for the nearest open models per llmrun Score and the best model that fits 24 GB. The highest score in each row is marked.

Board	Qwen2.5 7B Instructthis model	Qwen3.8 27Bbest in classQ4_K_M · 18 GB	Mistral Nemo Instruct 2407nearest by ScoreQ4_K_M · 11 GB	Gemma 3 4B ITnearest by ScoreQ4_K_M · 5.0 GB	Llama 3.1 8B Instructnearest by ScoreQ4_K_M · 7.9 GB
llmrun Score	65.1	122.1	65.4	64.0	62.5
GPQA Diamond	35.5%	—	29.9%	23.2%	27.0%
BALROG	7.8%	—	17.6%	—	15.1%
DTBench	47.7%	88.0%	48.6%	50.9%	50.9%
LMCA	6.4%	41.4%	—	2.8%	5.4%
AIME 2024/2025	2.5%	—	—	7.5%	1.7%
MMLU	72.9%	—	—	—	56.1%
Compare head to head		Compare	Compare	Compare	Compare

Scores aggregated from public benchmark sources (Epoch AI: 6 boards) — 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.