by Alibaba Cloud · founded 2023 · updated Jul 2026
Qwen is Alibaba Cloud's family of large language models spanning open-weight (Apache 2.0) and proprietary tiers, covering text, vision, audio, and coding use cases. It is available via the free Qwen Chat web/app, the Alibaba Cloud Model Studio API, or self-hosted from Hugging Face and ModelScope for teams needing on-prem control.
LLMs & AssistantsFreemiumEstablishedFrom Free (open-weight, self-hosted) / API from ~$0.05 per 1M input tokens
◆Developers wanting low-cost, self-hostable open-weight LLMs for coding and agentic workflows
◆Enterprises needing multilingual, multimodal models via a single unified API
◆Cost-sensitive production deployments seeking frontier-adjacent performance at a fraction of Western model pricing
Key integrations
Hugging FaceModelScopeAlibaba Cloud Model Studio (DashScope)vLLM / SGLang / OllamaOpenRouter / Together AI
Strengths
＋Aggressive price-to-performance with open-weight Apache 2.0 models that can be self-hosted at zero per-token cost
＋Strong, frequently-updated coding and agentic performance, including high SWE-bench Verified scores
＋Broad model catalogue spanning text, vision, audio, coding and embeddings under one API
Limitations
－Newer flagship tiers (Qwen3.6/3.7-Max) have shifted to closed weights, limiting self-hosting for the most capable models
－Some large models use the more restrictive Tongyi Qianwen License rather than Apache 2.0, with commercial caps tied to MAU thresholds
－Reported leadership departures in the Qwen team raise questions about continuity of the open-weight strategy
Evidence behind this score
Each score is built from the claims below. This is a snapshot, so check the current terms with the vendor before you commit to anything.
3 of 4 claims below are past their re-check window. Pricing, availability and security details move fast, so confirm those with the vendor.
Claim
Source
Checked
The entire Qwen3 series (0.6B through 235B-A22B) is Apache 2.0 licensed and available on Hugging Face, so it can be run locally at no per-token cost.· re-check
https://www.eesel.ai/blog/qwen-pricing
Jun 2026
Qwen 3.6-27B outperforms the much larger Qwen 3.5 (397B MoE) on agentic coding, scoring 77.2% on SWE-bench Verified.
https://theairankings.com/alibaba/qwen-3-6/
May 2026
Newer flagship models like Qwen 3.7-Max are API-only with no open weights, a departure from the historical open-release pattern.· re-check
Larger models such as Qwen3-235B-A22B use the Tongyi Qianwen License, which restricts commercial use above 100 million monthly active users and imposes redistribution conditions.· re-check
Four frontier-tier models shipped in three days — the race just isn't only closed models anymore
Grok 4.6 (Aug 12), Gemini 3.7 Flash (Aug 13), and open-weight Qwen3.8-27B and GLM-5.3 (both Aug 14) landed inside a single week. The closed-model pair both chose speed over a new flagship: Grok 4.6 is a post-training upgrade rather than a new base, and Gemini 3.7 Flash shipped as Google's fast workhorse while the actual flagship, Gemini 3.5 Pro, stays delayed. Meanwhile Alibaba and Z.ai both pushed open-weight models with frontier-adjacent coding scores that run on a single high-end GPU. The open-weight tier is no longer merely the cheap option; it is now the fast-moving one. A quarterly re-benchmarking cycle will not keep up with that.
GrokGeminiQwenxAI, Google, Alibaba Qwen and Z.ai official announcements; MarkTechPost, Axios and 9to5Google coverage
Releasehigh impactAug 3, 2026
Qwen3.8-Max: a frontier-scale model goes fully open, not just cheap
Alibaba's Qwen3.8-Max matches Sol and Opus 5 on API pricing today ($2/$6 per million tokens) and ships full weights within the week — the largest model yet to go open rather than staying API-locked. Every other open-weight lab now has to answer a 2.4T-parameter incumbent rather than merely a cheaper alternative to the closed frontier. Teams planning to self-host should watch the VRAM and serving requirements once weights land.
QwenAlibaba Qwen official announcement (qwen.ai); MarkTechPost and Dataconomy coverage