Qwen3 Coder Next
Last verified August 24, 2026 · Alibaba pricing ↗$0.120/1M input · $0.800/1M output · 262K context · Alibaba
Qwen3 Coder Next costs $0.120 per 1M input tokens and $0.800 per 1M output tokens from Alibaba, with a 262K-token context window. A typical chat reply (1,500 tokens in, 400 out) costs $0.0005, or $5.00 per 10,000 requests. On that workload it is cheaper than 136 of the 172 other buyable models tracked here.
Count Tokens
—
Tokens
—
Input Cost
—
Output Cost
Estimate Monthly Cost
Monthly Cost Estimator
Quick:
<$0.0001/mo
Pick a preset above or enter custom usage
What Qwen3 Coder Next Costs Per Request
At $0.120 per 1M input and $0.800 per 1M output tokens, before caching or batch discounts.
| Workload | Input / output tokens | Per request | Per 10,000 requests |
|---|---|---|---|
| Chat reply | 1,500 / 400 | $0.0005 | $5.00 |
| RAG answer over retrieved docs | 8,000 / 600 | $0.0014 | $14.40 |
| Coding-agent step | 30,000 / 2,000 | $0.0052 | $52.00 |
Qwen3 Coder Next vs Similarly Priced Models
Nearest buyable models by price, with the RAG answer above (8,000 in / 600 out) priced on each.
| Model | Input /1M | Output /1M | Context | RAG answer |
|---|---|---|---|---|
| Qwen3 Coder Next Alibaba | $0.120 | $0.800 | 262K | $0.0014 |
| Qwen3.8 Flash Alibaba | $0.150 | $0.470 | 1.0M | $0.0015 |
| Mercury 2 Inception Labs | $0.250 | $0.750 | 128K | $0.0024 |
| Hunyuan HY3 Preview Tencent | $0.178 | $0.595 | 262K | $0.0018 |
Alternatives to Qwen3 Coder Next
Pricing Details
Alibaba80B total / 3B active MoE coding specialist, Apache 2.0. Non-thinking mode only. The rate here is OpenRouter's cheapest host, which is where the money actually goes for this model — it was $0.11 input on August 3, 2026 and $0.12 on August 24, and it moves whenever a different host undercuts, so treat it as a marketplace floor rather than a list price. This row previously cited Alibaba's Model Studio page as its source while carrying OpenRouter's number, which Alibaba's page does not print: Alibaba DashScope sells the same model at $0.30/$1.50, and Bedrock, Hugging Face and Vercel at $0.50/$1.20. 256K native, extendable to 1M. SWE-Bench Verified 70.6 (SWE-Agent).
Input / 1M tokens
$0.12
Output / 1M tokens
$0.8
Context Window
262K
Max Output
66K
Price History
| Date | Input /1M | Output /1M | Change |
|---|---|---|---|
| 2026-08-24 | $0.110 → $0.120 | $0.800 → $0.800 | ↑ +9% |
Frequently Asked Questions
Common questions about Qwen3 Coder Next pricing and usage
How much does Qwen3 Coder Next cost?
Qwen3 Coder Next costs $0.12 per 1 million input tokens and $0.8 per 1 million output tokens. This pricing is set by Alibaba.
What tokenizer does Qwen3 Coder Next use?
Qwen3 Coder Next uses the cl100k_base tokenizer. This tokenizer has a vocabulary of ~100,000 tokens and is widely used across many models.
What is Qwen3 Coder Next's context window?
Qwen3 Coder Next supports a context window of 262K tokens (262,144 tokens). The maximum output length is 66K tokens.
How does Qwen3 Coder Next compare to other models?
On an 8,000-token-in, 600-token-out RAG answer, Qwen3 Coder Next costs $0.0014 per request. The nearest-priced alternatives cost: Qwen3.8 Flash (Alibaba) $0.0015, Mercury 2 (Inception Labs) $0.0024, Hunyuan HY3 Preview (Tencent) $0.0018. The table above lists their full rates and context windows.
How do I count tokens for Qwen3 Coder Next?
Use the token counter at the top of this page or visit the homepage calculator. Paste your text and it will show the exact token count and cost for Qwen3 Coder Next using the cl100k_base tokenizer.
Read More About Qwen3 Coder Next
Every big lab now ships a dedicated coding model. Their output prices run from $0.80 to $14 a million, the same job priced 17x apart.
July 24, 2026 · 8 min read
Qwen3 Coder Next is still $0.11 per million input. Kimi K2.6 costs 7x more, GPT-5.5 costs 41x more, and the benchmark gaps do not justify either.
May 14, 2026 · 10 min read