Qwen3 235B A22B · Alibaba (Qwen)

What does Qwen3 235B A22B cost to run in a coding agent?

Qwen3 235B A22B costs about $0.0096 per average coding-agent request on our measured traffic, at OpenRouter's listed prices of $0.09 per million input and $0.55 per million output tokens. It scores 12 on the Artificial Analysis Intelligence Index (an estimate), 45th of the 48 models we track. Source: Artificial Analysis (artificialanalysis.ai)

List prices: OpenRouter listed price, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. How we compute this

Run current Claude, GPT, Gemini and open models in your agent

Or let the router send each step to the cheapest model that can do it, on providers we vet for price, uptime and speed. Plans from $19 a month.

See plans

What Qwen3 235B costs per coding-agent request

Cost at list price	Qwen3 235B
Prompt under 100K tokens50% of requests, 87% of input cached	$0.0055
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.015
Prompt over 272K tokens11% of requests, 86% of input cached	n/a
Average request	$0.0096
100 requests	$0.96
Median session16 requests; a session ends after 10 idle minutes	$0.091
A month of regular usemedian of 20 users active on 5+ days	$13.05

Each logged request priced at the listed rates above for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request. 12% of requests were larger than Qwen3 235B's 262K context window and are left out of its figures.

Qwen3 235B benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	12 (estimate)	45 of 48
Output speedmedian	48 tokens/s	39 of 44
Time to first answer tokenincludes thinking time	1.3 s	3 of 44

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: Qwen3 235B A22B 2507 Instruct. Fetched October 13, 2026. Rank is among the models on this site, older versions included. Artificial Analysis runs Terminal-Bench 4.0 in its own setup, so its scores differ from the tbench.ai leaderboard.

What to watch for

Its 262K-token window is smaller than 12% of our coding-agent requests, which are left out of the figures above, so long agent sessions will hit the limit. No cache discount is listed, so the cached input that makes up most agent traffic costs the full $0.09 per million. The price is OpenRouter's listed rate on October 13, 2026, which can be a third-party host's; Alibaba's own API may charge differently.

About Qwen3 235B A22B

Qwen3 235B A22B (2507) is an open-weight mixture-of-experts model from July 2025: 235B parameters with about 22B active per token, so it is efficient to serve and can be self-hosted on suitable GPUs. It is now well behind current models: Artificial Analysis estimates 12 on its Intelligence Index and lists it as superseded. At OpenRouter's listed price there is no cache discount, and its 262K context is smaller than 11.7% of the coding-agent requests we measured.

Sources: Artificial Analysis

Qwen3 235B A22B FAQ

How much does Qwen3 235B A22B cost per coding-agent request?

$0.0096 on average at OpenRouter's listed price, $0.96 per 100 requests, and $0.091 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at OpenRouter's listed rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does Qwen3 235B A22B charge more for long prompts?

No. Qwen3 235B A22B bills the same rate per token up to its 262K-token context window, which is smaller than 12% of the coding-agent requests we measured.

How fast is Qwen3 235B A22B?

Artificial Analysis measures Qwen3 235B A22B 2507 Instruct at 48 output tokens per second, with 1.3 seconds to the first answer token, thinking time included. Source: Artificial Analysis (artificialanalysis.ai).

Is Qwen3 235B A22B good for AI agents?

Self-hosted setups that need a mid-size open-weight model. On our scoring, which is calibrated to public benchmarks, it rates 4.4/10 for agentic ability and 3.8/10 for output quality. Open weights, self-hostable.

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 or run it through a hosted provider such as OpenRouter. Check the license before commercial use.

What are the downsides of Qwen3 235B A22B?

The main trade-offs: Artificial Analysis estimates 12 on its Intelligence Index; No cache discount at OpenRouter's listed price; 262K context, smaller than 11.7% of the coding-agent requests we measured. It's strongest for self-hosted setups that need a mid-size open-weight model.