Qwen
Qwen
Language model · China ·Sep 2024

Qwen2.5-7B

Language Open (Restricted)
7.6B
Parameters
42
Benchmarks

About Qwen2.5-7B

Qwen2.5-7B is an AI model developed by Qwen, in the language category, released in 2024, made available as an open-weights model with 7.6B params.

On this page you'll find Qwen2.5-7B's full specifications, including 42 benchmark results. Review provider pricing and benchmark scores below, or compare Qwen2.5-7B head-to-head with other language models.

Links

Run Qwen2.5-7B locally — what to buy

Share this build ↗

Sized to what Qwen2.5-7B actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Software support

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Community

Would you run Qwen2.5-7B again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

Qwen2.5-7B benchmark scores
Benchmark Category Score Variant
HumanEval code 93.35% instruct-cited-granite-3-2
HumanEval+ code 89.91% instruct-cited-granite-3-2
AlpacaEval 2 general 31.7% instruct-cited-exaone35-7b
AlpacaEval 2 general 30.34% instruct-cited-granite-3-2
ArenaHard general 55.5% cited-phi4-mini
ArenaHard general 48.9% instruct-cited-exaone35-7b
ArenaHard general 25.44% instruct-cited-granite-3-2
IFEval general 74.9% instruct-cited-granite-3-2
IFEval general 74.7% instruct-cited-falcon3
LiveBench general 35.6 instruct-cited-exaone35-7b
MMLU general 74.4% base-cited-olmo2
MMLU general 73.5% instruct-5-shot-cited-falcon3
MMLU general 72.6% cited-phi4-mini
MMLU-Pro general 56.2% cited-phi4-mini
MMLU-Pro general 45.8% base-cited-olmo2
MMLU-Pro general 43.1% instruct-5-shot-cited-falcon3
MMMLU general 64.4% cited-phi4-mini
AGIEval knowledge 63.7 base-cited-olmo2
ARC-Challenge knowledge 90.1 cited-phi4-mini
ARC-Challenge knowledge 89.5 base-cited-olmo2
ARC-Challenge knowledge 57.8 instruct-25-shot-cited-falcon3
TriviaQA knowledge 69.4 base-cited-olmo2
GSM8K math 88.7% cited-phi4-mini
GSM8K math 84.46% instruct-cited-granite-3-2
GSM8K math 81.5% base-cited-olmo2
GSM8K math 72% instruct-5-shot-cited-falcon3
MATH math 60.4% cited-phi4-mini
MGSM math 64.5 cited-phi4-mini
BIG-Bench Hard reasoning 72.4% cited-phi4-mini
BIG-Bench Hard reasoning 70.4% instruct-cited-granite-3-2
BIG-Bench Hard reasoning 54.1% instruct-3-shot-cited-falcon3
DROP reasoning 55.8% base-cited-olmo2
DROP reasoning 54.71% instruct-cited-granite-3-2
GPQA Diamond reasoning 32% instruct-0-shot-cited-falcon3
HellaSwag reasoning 89.7 base-cited-olmo2
HellaSwag reasoning 80 cited-phi4-mini
PIQA reasoning 76.2 cited-phi4-mini
PIQA reasoning 73.7 instruct-0-shot-cited-falcon3
Winogrande reasoning 74.2 base-cited-olmo2
Winogrande reasoning 71.1 cited-phi4-mini
TruthfulQA safety 69.4% cited-phi4-mini
TruthfulQA safety 63.06% instruct-cited-granite-3-2

More models like Qwen2.5-7B

Frequently asked questions

What is Qwen2.5-7B?

Qwen2.5-7B is an AI model developed by Qwen, in the language category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Qwen2.5-7B?

Qwen2.5-7B was developed by Qwen and released in 2024.

Is Qwen2.5-7B open source or proprietary?

Qwen2.5-7B ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

How much does Qwen2.5-7B cost?

Pricing for Qwen2.5-7B depends on the provider. See the providers table on this page for the latest API rates.

How does Qwen2.5-7B perform on benchmarks?

Qwen2.5-7B is benchmarked across 42 evaluations on GenAIList, including HumanEval (93.35%). See the full benchmark table below and compare it with other models.

No videos yet. Know a good one? Add it below.

Know a great video about Qwen2.5-7B?