Qwen3.6 Max Preview
by Alibaba (Qwen) · qwen/qwen3.6-max-preview · #233 cheapest of 325 paid models
Prices updated Jul 22, 2026, 1:31 AM UTC · refreshed hourly
Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters. It is optimized for agentic coding, tool use, and...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$1.04
per 1M tokens
Output
$6.24
per 1M tokens
Blended (3:1)
$2.34
3 input : 1 output
Cache read
—
per 1M cached tokens
Cache write
$1.30
per 1M tokens
Cache & batch economics
Effective prices for Qwen3.6 Max Preview (no cache-read rate published — hits billed as normal input).
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $2.34 | $0.1123 | — |
| 50% | $2.34 | $0.1123 | −0% |
| 90% | $2.34 | $0.1123 | −0% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.002392 | $2.39 |
| Document summary | 8,000 | 1,000 | $0.0146 | $14.56 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.0437 | $43.68 |
| Long-context analysis | 150,000 | 5,000 | $0.1872 | $187.20 |
Performance
Intelligence Index
40
AA composite quality
Coding Index
—
Math Index
—
Agentic Index
—
Output speed
46 tok/s
median
Time to first token
2.10s
median
Time to first answer
45.92s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| GPQA Diamond | 88.8% |
| Humanity's Last Exam | 28.9% |
| SciCode | 46.9% |
| IFBench | 76.6% |
| AA-LCR (Long Context Reasoning) | 69.7% |
| Terminal-Bench Hard | 43.9% |
| τ²-Bench (Telecom) | 95.9% |
Specs
Context window
262K
tokens
Max output
66K
tokens
Input modalities
text
Tokenizer
Qwen