Log in
← All models

Qwen3 235B A22B

Budget

A large open-weight MoE you can self-host — capable and extremely cheap to run hosted.

Alibaba (Qwen)
Maker
82/100
Overall
99
Value score
$0.09
Blended /1M
262K
Context
Open
Weights
Pricing: $0.09 in · $0.10 out / 1M tokens · synced from OpenRouter · updated 2026-06-22

Scorecard

Output Quality8.0/10 · 72% community
Agentic Ability8.0/10 · 72% community
Speed8.2/10 · 73% community
Value for $10.0/10 · 82% community
Reliability8.8/10 · 76% community

Qwen3 235B A22B is a large open-weight mixture-of-experts model: only ~22B parameters active per token, so it's cheap to serve while staying capable. Open weights make it a go-to for self-hosted, privacy-sensitive agent setups.

Strengths
  • Open weights, self-hostable
  • Very cheap hosted
  • Efficient MoE
Trade-offs
  • Behind frontier models on quality
  • Needs real GPUs to self-host well

Agents that run Qwen3 235B well

Self-hosted or ultra-cheap open-weight agents.

Best model for ClineBest model for ContinueBest model for OpenClawBest model for HermesBest model for OpenCode

Qwen3 235B A22B FAQ

Is Qwen3 235B A22B good for AI agents?
Self-hosted or ultra-cheap open-weight agents. On our editorial scoring it rates 8.0/10 for agentic ability and 8.0/10 for output quality. Open weights, self-hostable.
How much does Qwen3 235B A22B cost?
Via OpenRouter, Qwen3 235B A22B is priced at $0.09 in · $0.10 out / 1M tokens — a blended rate of about $0.09 per 1M tokens for typical input-heavy agent use.
What is Qwen3 235B A22B's context window?
Qwen3 235B A22B has a 262K-token context window. Alibaba (Qwen) is the maker.
Is Qwen3 235B A22B open source?
Yes — Qwen3 235B A22B ships with open weights, so you can self-host it for privacy or run it through a hosted provider like OpenRouter. That openness is a big reason it's popular in open agent stacks.
What are the downsides of Qwen3 235B A22B?
The main trade-offs: behind frontier models on quality; needs real gpus to self-host well. It's strongest for self-hosted or ultra-cheap open-weight agents.

Qwen3 235B compared

Qwen3 235B vs Haiku 4.5Qwen3 235B vs DeepSeek V4 FlashQwen3 235B vs GPT-5.4 MiniQwen3 235B vs Flash LiteQwen3 235B vs Llama 4 MaverickQwen3 235B vs CodestralQwen3 235B vs Qwen3.7 MaxQwen3 235B vs Qwen3.7 Plus

The model is half the story — the agent is the other half

The model picks the moves; the agent runs the loop, the tools, and the guardrails. Once you've chosen a model, see which agent gets the most out of it.

Compare AI agents →