Alibaba Cloud Cuts Qwen4-Max Singapore Prices 25% Versus Qwen3.8-Max
Alibaba Cloud launched Qwen4-Max on November 16, 2026, with Singapore prices 25% below Qwen3.8-Max: $1.50 per million input and $4.50 per million output tokens.
By Amara Osei
3 min read

What's News
- Qwen4-Max lists at $1.50/$4.50 per million input/output tokens in Singapore, 25% below Qwen3.8-Max.
- Alibaba Cloud released the model through Model Studio on November 16, 2026.
- Thinking and the final answer share a 262,144-token output allowance.
- Cache reads cost $0.075 per million tokens; fine-tuning is unsupported in all regions.
- The Qwen4-Max-Preview retires on November 25, 2026.
Alibaba Cloud cut Singapore list prices for Qwen4-Max by 25% versus Qwen3.8-Max, setting the new flagship at $1.50 per million input tokens and $4.50 per million output tokens. Qwen3.8-Max carried $2 and $6 respectively. The company released the model through Model Studio on November 16, 2026.
The release pairs the price cut with a one-million-token free quota in Singapore, valid for 90 days from the latest of Model Studio activation, model release or application approval. Qwen4-Max is a multimodal model with a one-million-token context window.
What can developers actually build with it?
The model documentation describes text, image and video input with text output. Supported features include streaming, web search, parallel tool calling, function calling, prefix completion and structured JSON output. Developers can integrate through OpenAI Chat Completions, OpenAI Responses, Anthropic Messages or Alibaba Cloud's own DashScope APIs — several established routes for coding and agent software.
Alibaba Cloud built Qwen4-Max for coding, office productivity, frontend design and complex agentic tasks. Thinking runs by default at high reasoning effort, with low, high and xhigh settings available.
The token economics carry a catch. The default thinking budget is 131,072 tokens, which matches the maximum non-thinking output. Thinking and the final answer share a single 262,144-token allowance, so deliberation consumes part of the same output budget as the response itself.
How does caching change the bill?
The Qwen4 caching guide allows identical input prefixes — system prompts, tool descriptions, reference documents — to be reused across requests from the same user. An explicit cache prefix must contain at least 1,024 tokens, with five-minute or one-hour lifetimes.
- Explicit breakpoints guarantee deterministic hits for identical prefixes within the validity window.
- Implicit caching spots common prefixes automatically but doesn't guarantee hits.
- Automatic caching can place a breakpoint at the end of the input, letting it advance as a conversation grows.
Cache reads in Singapore cost $0.075 per million tokens under either approach. Creating an explicit cache costs $1.875 per million tokens for five minutes and $3 for one hour. Only input tokens get the discount; output tokens stay at full price. For agents that repeatedly send long instructions or documents, prefix stability and cache lifetime determine whether lower input charges actually follow the reduced list rates.
Where is it available, and what's off-limits?
Qwen4-Max is listed in Beijing, Singapore, Frankfurt, Virginia, Tokyo and Hong Kong. Batch inference runs only in Beijing. Fine-tuning and custom model deployment are unsupported across every documented region. That leaves API configuration, reasoning settings and caching as the main levers for controlling the managed model.
The full model also ships in Singapore's Personal Edition Token Plan, billed through Credits. The plan supports interactive tools including Claude Code, Cursor, Qwen Code, Qoder and OpenClaw. But the subscription is limited to interactive coding and agent-tool use. Automation scripts, custom application backends and non-interactive batch calls are prohibited. Pay-as-you-go API calls remain the alternative route.
What came before, and what happens next?
The full release follows Qwen4-Max-Preview, which went live at 10:00 on November 11, 2026 (UTC+8) exclusively through Token Plan. The preview runs through November 25, 2026, after which Alibaba Cloud retires the preview model.
The lower rates and caching options target input costs for long-running agent workloads. The final bill still hinges on output length, reasoning depth and cache reuse — and with fine-tuning and custom deployment off the table, those usage controls plus the choice between API and subscription access will decide the release's practical impact.
Source: unite.ai
More from Amara Osei
Show full bio
Senior reporter covering companies, deals and strategy in entrepreneurship, startups and venture funding at Business Bearings.
2,136 articles