TECH Signal 390

Alibaba prices Qwen3.8-Max at $2 per 1M input tokens and $6 per 1M output tokens via its API, below Kimi K3's $3/1M input tokens and $15/1M output tokens (Henry Siu/The Information)

Techmeme 6 hours ago 43 min read ↗ Original

WHY IT MATTERS

Engineers now have a cheaper option for high-volume inference workloads. The price gap—especially on output tokens—can shift cost-sensitive projects toward Qwen3.8-Max. No other feeds corroborate the pricing or performance claims, so independent validation is still needed.

Written by elseif from the cluster below · every claim links back to a source

The three things worth knowing

Input tokens cost $2 per million, output tokens $6 per million—both below Kimi K3’s published rates.

The model is accessible via API; weights for a smaller variant will be released next week.

Pricing alone does not guarantee performance; benchmarks cited in the material are not detailed here.

THE CLUSTER

Same story, 1 feed.

ORDERED BY FIRST SEEN

Techmeme Alibaba prices Qwen3.8-Max at $2 per 1M input tokens and $6 per 1M output tokens via its API, below Kimi K3's $3/1M input tokens and $15/1M output tokens (Henry Siu/The Information) 6 hours Open ↗