Qwen3.8-Max
Alibaba · Lanzado jul 2026
Buque insignia MoE multimodal de 2.4 billones de parámetros. Todavía no en GA: sin precios de API por token, sin tabla de benchmarks, sin pesos abiertos publicados. Entrada desde $6/mes mediante Token Plan Lite.
¿Es adecuada para ti?
Bueno para
- Published per-token rate: $2 input / $6 output per 1M on Alibaba Cloud Model Studio (Singapore), one tier up to 1M tokens, with context-caching discount
- Native multimodal input — Alibaba Cloud's model list places it under both text generation and image and video understanding
- OpenAI and Anthropic API-compatible — works with Cursor, Claude Code, OpenCode, and Codex
No recomendado para
- Choosing on benchmarks — we found no published benchmark table or model card, and the 'second only to Fable 5' claim is Alibaba's own self-assessment
- Self-hosting — 2.4T parameters would need datacenter-scale infrastructure, and we could not confirm an open-weight release
- Regulated environments requiring auditable model cards and verified benchmark data
Precios
Entrada
$2.00 / 1M tokens
Salida
$6.00 / 1M tokens
Contexto
Model Studio (Singapore); one tier up to 1M tokens
Aún no hay cambios de veredicto
El reloj corre desde el primer día: los cambios aparecerán aquí a medida que evolucione nuestro veredicto.
Fuentes
- Alibaba Cloud Model Studio pricingoct 2026
- Kie.ai — What Is Qwen3.8-Max? Alibaba's 2.4T Flagshipago 2026
- MarkTechPost — Alibaba Previews Qwen3.8-Maxago 2026
- Alibaba Cloud Model Studio: Modelssept 2026
- Coursiv — Qwen 3.8: Specs, Pricing, Access & Benchmarksago 2026
- Lumien — Alibaba's Qwen3.8-Max: 2.4 Trillion Parameters, No Benchmarks Yetago 2026
Registro de verificación
- Precios· Sin cambios
Agente automatizado
- Precios· Sin cambios
Agente automatizado
- Precios· Actualizado
Agente automatizado
Stored pricing_input/output already read $2/$6 but the profile still said no per-token pricing was published and the rating was pending. Alibaba Cloud lists Qwen3.8-Max at $2/$6 per 1M (Singapore, one tier up to 1M). Strengths, audiences, pricing_context, pricing_url and sources refreshed. Verdict pending -> conditional staged for publish on draft qwen3-8-max-conditional-published-pricing. The stored verdict_summary still says no per-token pricing until that publish lands.
- Precios· Sin cambios
Agente automatizado
The International rate card lists qwen3.8-max at $2 input / $6 output per 1M (tier 0 to 1M tokens), which matches the stored scalars. Flag for a verdict review: the stored summary and rating (pending) still say there is no per-token pricing, no GA and open weights 'promised week of Aug 10', and the rate card now contradicts the pricing part. A rating move needs an announcing post and GA evidence, so nothing was written this run.
- Precios· Sin cambios
Agente automatizado