Qwen

Qwen3-235B-A22B

Name: Qwen3-235B-A22B
Author: Qwen

Qwen3-235B-A22B is a Qwen model with 33K context, Not verified in / Not verified out per million tokens, last verified October 25, 2026. Open weights: yes. Status: Legacy. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

Qwen3-235B-A22B is Alibaba's open-weight MoE with 235B total and 22B active parameters. The original checkpoint has 32,768 native context, extended to 131,072 with YaRN.

Recorded profile verification October 25, 2026Individual provider fields may change

AI model specification details
Specification	Qwen3-235B-A22B
Provider	Qwen
Tier	Balanced
Context window	33K
Max output	Not verifiedUnverified
Input / 1M tokens	Not verifiedUnverified
Output / 1M tokens	Not verifiedUnverified
Weights	Open
Parameters	235B total / 22B active (MoE)
Reasoning levels	Not verifiedUnverified
Modalities	text
License	Apache 2.0
Released	April 29, 2025

Pricing tiers: No current first-party tariff for the original Qwen3-235B-A22B checkpoint was recovered from the reviewed Model Studio list. Its former hosted quote is not a current offer. Apache-2.0 weights remain available for separately costed self-hosting or a specifically verified provider. Native 32,768 context extends to 131,072 with YaRN; recommended 32,768 generation is not a hard cap.

Best for

Open-weight deployments
Self-hosting
Multilingual

Watch out

Original native hybrid checkpoint: thinking/non-thinking templates, 235B total/22B active and 32,768 native context. Do not replace it with a later Instruct/Thinking checkpoint or transfer a hosted 256K contract. Current first-party endpoint identity, price and hard output cap remain unverified. Absence from a price list does not prove universal retirement; Apache weights remain available.

Source receipts

Catalog figures for this model were checked against the following sources.

Current scoped source review: exact publisher model, tariff, licence or lifecycle evidence (accessed 2026-10-25)
Current scoped source review: exact publisher model, tariff, licence or lifecycle evidence (accessed 2026-10-25)
Current scoped source review: exact publisher model, tariff, licence or lifecycle evidence (accessed 2026-10-25)
Current scoped source review: exact publisher model, tariff, licence or lifecycle evidence (accessed 2026-10-25)
Current scoped source review: exact publisher model, tariff, licence or lifecycle evidence (accessed 2026-10-25)

Common questions

Qwen3-235B-A22B

Answered from the verified figures on this page rather than general guidance.

What is Qwen3-235B-A22B's context window?

Qwen3-235B-A22B accepts about 33K tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.

What is Qwen3-235B-A22B best for?

Qwen3-235B-A22B is a balanced tier from Qwen. It suits open-weight deployments, self-hosting, multilingual. Original native hybrid checkpoint: thinking/non-thinking templates, 235B total/22B active and 32,768 native context. Do not replace it with a later Instruct/Thinking checkpoint or transfer a hosted 256K contract. Current first-party endpoint identity, price and hard output cap remain unverified. Absence from a price list does not prove universal retirement; Apache weights remain available.

Can I self-host Qwen3-235B-A22B?

Qwen3-235B-A22B publishes open weights, but self-hosting depends on the licence, hardware footprint, quantisation quality, and serving stack. A hosted API is often cheaper until you have measured throughput and concurrency on your own hardware.

Model comparison catalog Benchmark library