Qwen 3
Product status: Status not recorded
No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
No status check date recorded
Alibaba's original 2025 Qwen3 open-weight family with thinking/non-thinking modes; distinct from Qwen Chat and later Qwen generations. No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
Product preview
Decision notes
Official-document research
Based on the cited documentation. We did not use the product, purchase a plan, or run quality, speed or reliability tests in this review.
Consider it for
- Teams maintaining and evaluating a model stack
Choose another approach if
- Assuming family name means current Chat/API entitlements
Pricing and conditions
Current terms unconfirmed
The original Apache-2.0 release has no software-license royalty, but inference hardware, memory, storage, deployment and any provider fees are separate. The current Model Studio page includes later generations and region-specific token/cache/discount conditions; an exact current original-Qwen3 quote was not verified. Currency, tax, region, input/output unit, quota, billing term and overage must be matched to the selected model and provider. Later-model rates and Singapore-only promotional quotas are not copied into this entry.
Pricing source read:
Pricing or terms sourceCode, model and service licensing
The launch identifies the original released family as Apache-2.0, and the original 235B-A22B weight repository includes that license. Commercial use and derivatives are subject to notice, modification-marking, license and applicable patent terms; trademarks and separately bundled components remain separate. This is not a Qwen Chat subscription, Model Studio contract or grant in all future Qwen weights. Inputs, generated-material rights and any external provider's training/retention conditions require separate review; local deployment alone does not prove network isolation.
Basis and limits of the recorded score
The existing 4.6 score is retained; its original scoring evidence is unavailable. This documentation review did not measure performance or compare products and did not re-score it.
Overview
This entry covers the Qwen3 family announced on 2025-04-29: two MoE and six dense releases, including original Qwen3-235B-A22B. It does not merge that checkpoint with a later dated suffix or the hosted Qwen Chat application. The card documents hard thinking-mode control and prompt-level soft switches, with generation-setting guidance. Choose a checkpoint, context configuration and deployment stack, then evaluate the task; the launch benchmarks and language-coverage figures are not independent proof of your results.
Sources
The factual claims in this review were read from the pages below. Each date is when an editor last opened the source, not when the product changed.
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor pricing page Checked 2026-10-15
Pros / Cons
Pros
- Model sizes and deployment examples
- Explicit thinking-mode settings
Cons
- Checkpoint/version selection matters
- Resource and output evaluation needed
Who it's for
Self-managed text and code experiments
Task-specific checkpoint evaluation
Bottom line
Use the exact original checkpoint and its license; confirm cloud access and charges independently.
Key Features
Tags
Compare Qwen 3 with alternatives
Frequently asked questions
Is Qwen 3 currently available?
No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
What is Qwen 3?
Qwen 3 is a Conversational AI tool developed by Alibaba. Alibaba's original 2025 Qwen3 open-weight family with thinking/non-thinking modes; distinct from Qwen Chat and later Qwen generations.
What costs and terms should I check for Qwen 3?
The original Apache-2.0 release has no software-license royalty, but inference hardware, memory, storage, deployment and any provider fees are separate. The current Model Studio page includes later generations and region-specific token/cache/discount conditions; an exact current original-Qwen3 quote was not verified. Currency, tax, region, input/output unit, quota, billing term and overage must be matched to the selected model and provider. Later-model rates and Singapore-only promotional quotas are not copied into this entry.
Is Qwen 3 open source?
The launch identifies the original released family as Apache-2.0, and the original 235B-A22B weight repository includes that license. Commercial use and derivatives are subject to notice, modification-marking, license and applicable patent terms; trademarks and separately bundled components remain separate. This is not a Qwen Chat subscription, Model Studio contract or grant in all future Qwen weights. Inputs, generated-material rights and any external provider's training/retention conditions require separate review; local deployment alone does not prove network isolation.
What are the best alternatives to Qwen 3?
Notable Conversational AI alternatives include ChatGPT, Claude, Gemini. Browse all 67 tools in our Conversational AI category for a full overview.
Was this listing useful?
Basic Info
- Pricing
- Free
The original Apache-2.0 release has no software-license royalty, but inference hardware, memory, storage, deployment and any provider fees are separate. The current Model Studio page includes later generations and region-specific token/cache/discount conditions; an exact current original-Qwen3 quote was not verified. Currency, tax, region, input/output unit, quota, billing term and overage must be matched to the selected model and provider. Later-model rates and Singapore-only promotional quotas are not copied into this entry.
- Editorial score
- ★ 4.6 / 5 How we score
- Platforms
- Self-hostedModel weights
- Last verified
- 2026-10-12
