Qwen2.5-32B-Instruct
Mid-size 32B instruction-tuned model balancing capability and cost.
About
Qwen2.5-32B-Instruct is the 32.5 billion parameter instruction-tuned member of Alibaba's Qwen2.5 family, released in September 2024. It carries the same 128K context window, 29+ language support, and improved instruction following as its larger siblings, in a package that fits on a single high-memory GPU node.
The 32B size became a popular sweet spot for self-hosted chatbots and agent backends, offering near-flagship quality at roughly half the serving cost of the 72B variant, under the Apache 2.0 license.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 131; Trending score 84/100 from researcher assessment.