Qwen4-Max vs. Claude Opus 5.5: The cheaper model costs you somewhere else.
Qwen4-Max, made available in preview to Token Plan subscribers on November 11 and generally released on November 16, is quite The post Qwen4-Max vs. Claude Opus 5.5: The cheaper model costs you somewhere else. appeared first on The New Stack .
Alibaba's latest model Qwen4-Max has been generating a lot of buzz since its preview release on November 11 and general release on November 16. Independent testing firm Artificial Analysis conducted a standard test suite on the new model and found that it generated 127 million output tokens, roughly twice the median for comparable models. In contrast, Claude Opus 5.5 generated 218 million output tokens with default effort settings, but Opus's default effort setting is medium.
The cost difference between the two models is significant. Qwen4-Max costs $1.50 per million input tokens and $4.50 per million output tokens, while Claude Opus 5.5 costs $4 per million input tokens and $20 per million output tokens. Alibaba also charges separately for cached input and cache writes. Qwen4-Max's cache reads cost $0.075 per million tokens, while Opus 5.5 cache reads cost $0.20 per million tokens. These costs can add up quickly, so it's important to consider them when choosing a model.
While Qwen4-Max generates fewer output tokens than Opus 5.5, it also takes more time and turns to complete tasks. In the tests run by the reporter, Qwen4-Max scored 71 out of 76 points in the multi-app workflow test, compared to Opus 5.5's 76 points. Qwen4-Max also scored 59 out of 60 points in the incident investigation test, compared to Opus 5.5's 58 points.
Finally, in the agent bug fix test, Qwen4-Max scored 40 out of 47 points, compared to Opus 5.5's 45 points. Despite these differences, Qwen4-Max still offers a more cost-effective option for many users.
Written by urgent.news from The New Stack's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.