Qwen2.5 (7B + 72B): benchmark scores, price and ranking

Qwen2.5 (7B + 72B) from Qwen ranks #374 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.2), based on 2 benchmark results captured up to 2026-10-13.

Organisation: Qwen
Type: Frontier model
Open weights: No
Aggregate rank: #374 of 743 (score 48.2)

Across the 2 benchmarks where at least ten models are scored, Qwen2.5 (7B + 72B) is in the top 10% on 0 and in the top half on 0.

Qwen2.5 (7B + 72B) benchmark results

Qwen2.5 (7B + 72B) benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
SWE-bench Lite	24.7%	#11 of 20	SWE-bench Lite leaderboard	2026-10-13
SWE-bench Verified	32.8%	#59 of 73	SWE-bench Verified leaderboard	2026-10-13

About Qwen2.5 (7B + 72B)

Qwen2.5 (7B + 72B) is a frontier model from Qwen. No API price is listed for it yet. We hold 2 benchmark results for it across 1 area, and none is in the top tenth of its board.

Where Qwen2.5 (7B + 72B) stands

On our aggregate score, Qwen2.5 (7B + 72B) is #374 of 743 ranked models, around the middle of the table. Its score is 48.2; the scale is centred on 50, so it sits 1.8 points below the typical model. That beats 50% of the ranked field. Among the 23 ranked Qwen models it is #12. The next one up is Qwen2.5 Coder 32B Instruct at 48.3. The next one down is Qwen2.5-VL-72B-Instruct at 47.6.

Qwen2.5 (7B + 72B), benchmark by benchmark

Qwen2.5 (7B + 72B) lands below the middle on SWE-bench Lite: 24.7%, #11 of 20, with 50% of the field scoring higher. Claude 4 Sonnet leads at 56.7%; the closest result above is GPT-4o at 24.7%. It covers real GitHub issues resolved. The median score on this board is also 24.7%. Source: SWE-bench Lite leaderboard, captured 2026-10-13.

SWE-bench Verified is a weak spot. Qwen2.5 (7B + 72B) scores 32.8%, #59 of 73, behind 79% of the models measured. The leader, Claude Opus 4.7 (max), reaches 83.5%. It covers real GitHub issues resolved. The board median is 56.4%, so Qwen2.5 (7B + 72B) is 23.6 points below it. Source: SWE-bench Verified leaderboard, captured 2026-10-13.

Strengths and weaknesses

Software engineering: uneven, ahead of 32% of the field on average across 2 benchmarks. Its best board here is SWE-bench Lite (#11 of 20) and its weakest is SWE-bench Verified (#59 of 73).

Reading these numbers

The 2 scores on this page come from 2 sources: SWE-bench Lite leaderboard (1), SWE-bench Verified leaderboard (1). 2 of them come from an official leaderboard and 0 from datasets or third-party runs. With 2 results the aggregate rank can move a lot when one new score arrives.