Qwen3.8-Max — Specifications
Compilato da Mustafa Ihsan dalla documentazione pubblicata del fornitore · Ultimo aggiornamento
| Sviluppatore | Alibaba |
|---|---|
| Tipo | LLM (architettura MoE, ragionamento, multimodale) |
| Modalità | Text, Image, Video → Text |
| Parametri | 2.4T total (MoE) |
| Finestra contestuale | 1 milione |
| Output massimo | 128K |
| Licenza | Proprietary (hosted) |
| Pesi aperti | No |
| Pubblicato | 2026-08 |
| Prezzo dell’input | $2.00 /1M |
| Prezzo dell’output | $6.00 /1M |
| Provider API | Alibaba Cloud Model Studio, Qwen Cloud |
What is Qwen3.8-Max?
Qwen3.8-Max is Alibaba’s flagship Qwen model, launched on Alibaba Cloud Model Studio (International) on 2 August 2026. Alibaba lo descrive as a 2.4-trillion-parameter mixture-of-experts with a major leap in coding and office productivity, built to handle professional tasks in law, finance and design, with native understanding of images and video. Qwen calls it the official version of its open-weight Qwen3.8-2.4T-A95B (2.4T total, 95B active parameters), with extra features such as vision input, a non-thinking mode, a 1M-token context by default and built-in tools. An upgraded snapshot, qwen3.8-max-0902, followed on 2 September 2026 at the same price.
In Model Studio’s US and International deployment scopes it costs $2.00 per million input tokens and $6.00 per million output tokens, with implicit cache hits at $0.25 per million. Model Studio’s Global deployment scope lists it at $1.65 / $4.951, and Qwen Cloud, the official Qwen API service, charges $2.00 / $6.00. The context window is 1,000,000 tokens, with up to 131,072 output tokens.
It succeeds the April 2025 Qwen3 generation at the top of Alibaba’s range; the open Qwen3 235B-A22B remains the self-hostable flagship of that family, with a 32K native context. For Alibaba’s smaller open-weight Qwen3.8 model, see Qwen3.8-27B. Background on the family is in our Spiegazione di Qwen di Alibaba guide, and the Calcolatore dei costi delle API per l'IA estimates a monthly bill.
Qwen3.8-Max pricing: API cost per 1M tokens
| Input (per ogni milione di token) | $2.00 |
|---|---|
| Output (per ogni milione di token) | $6.00 |
| Rapporto output/input | 3× |
| Combinato (4:1 in:out) | $2.80 per 1 milione di token |
What Qwen3.8-Max costs per month
Spesa mensile reale con un mix input-to-output 4:1 — il rapporto effettivamente prodotto da un tipico carico di lavoro basato su chat o RAG.
| Carico di lavoro | Token/mese | Costo/mese |
|---|---|---|
| Progetto secondario | 1 milione in ingresso / 0,25 milioni in uscita | $3.50 |
| Piccolo team | 20 milioni in ingresso / 5 milioni in uscita | $70 |
| Produzione | 200 milioni in ingresso / 50 milioni in uscita | $700 |
Calcola i tuoi numeri personalizzati nel Calcolatore dei costi delle API per l'IA.
Cheaper alternatives to Qwen3.8-Max
| Modello | Costo combinato ($/1 milione) | Risparmi |
|---|---|---|
| Llama 3.1 8B aperta | $0.0240 | 99% più economico |
| Mistral NeMo 12B aperta | $0.0292 | 99% più economico |
| Gemma 3 4B aperta | $0.0600 | 98% cheaper |
Domande frequenti
How much does Qwen3.8-Max cost per 1M tokens?
Qwen3.8-Max costs $2.00 per 1M input tokens and $6.00 per 1M output tokens. At a typical 4:1 input-to-output mix that blends to about $2.80 per 1M tokens.
How much does Qwen3.8-Max cost per month?
A small-team workload of 20M input and 5M output tokens a month costs about $70 on Qwen3.8-Max. A side project (1M in / 0.25M out) costs roughly $3.50.
What is a cheaper alternative to Qwen3.8-Max?
Llama 3.1 8B is the strongest cheaper option in our database at $0.0240 per 1M blended — about 99% less than Qwen3.8-Max. It is also open-weight, so self-hosting is an option.
Can I run Qwen3.8-Max locally?
No. Qwen3.8-Max is a closed, API-only model — the weights are not released, so it cannot be self-hosted.
Why does Qwen3.8-Max charge more for output than input?
Output tokens are generated one at a time and cannot be batched the way a prompt can, so they cost the provider more to serve. Qwen3.8-Max charges 3× more for output, which is why prompt-heavy workloads are far cheaper to run than generation-heavy ones.
I prezzi indicati sono le tariffe ufficiali pubblicate per l'API principale del modello e vengono aggiornati man mano che i fornitori li modificano. Sconti per volumi elevati, elaborazione batch e input memorizzati nella cache non sono inclusi. Confronta tutti i modelli fianco a fianco nel database dei modelli AI o il Classifica LLM.