Summary
According to a post on X by @QwenDevs, Alibaba has launched Qwen4-Max, a model built on a new MoE architecture with native vision and a 1M-token context. Alibaba says it beats its previous Qwen3.8-Max while cutting average cost per task by more than 60%. API pricing is listed at $1.50 per million input tokens and $4.50 per million output tokens. Artificial Analysis scores it 43 on its Intelligence Index, three points behind Kimi K3.1 Pro, which the post places at about 11% lower cost per task.
AI-assisted summary based on the listed source.
What happened
Alibaba says Qwen4-Max, on a new MoE architecture, beats Qwen3.8-Max while cutting average cost per task by more than 60%, with API pricing of $1.50 per million input and $4.50 per million output tokens; Artificial Analysis scores it 43 on its Intelligence Index, 3 behind Kimi K3.1 Pro at about 11% lower cost per...
Why it matters
Developers and businesses comparing models on price may find these pricing and cost-per-task figures relevant to their workloads. The cost comparisons come from Alibaba's own claims, so independent testing will matter for judging them. Watch for the promised smaller open Qwen4 models, which could matter for teams that want to run a model themselves.
Signal Intelligence
Signal Strength 94%
Technical label WATCH
Public Interest 71
Category PRODUCT UPDATE
Reader Depth TECHNICAL
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
Public Interest components
Recognizable Entity Score 68
Practical Impact Score 74
Novelty Interest Score 94
Consequence Score 50
Curiosity Score 52
Shareability Score 73
Why this is here
VQV surfaced this signal because it is recent, relevant to Consumer AI, connected to Scobleizer News Pulse (X).