Cheaper Inference
Model record / Alibaba
qwen3-max-thinking
Access qwen3-max-thinking from Alibaba through one OpenAI-compatible API with live market-linked pricing and usage-based billing.
Text generationText inputReasoningStreaming35% below list
What you pay
Current catalog rates.
Input / 1M tokens$1.20 list$0.78
Output / 1M tokens$6.00 list$3.90
Cached input / 1M tokens$0.39
Current discount35%
Requests above 32,000 input tokens use $1.56 input and $7.80 output per 1M tokens.
Observed reliability
0
Successful provider-reaching requests
Output speed
—
Median tokens per second over sustained generations
TTFT
—
p50 · p90 not available · n 0
Latency
—
p50 · p90 not available · n 0
Evidence strength
—
At a glance
- Provider
- Alibaba
- Model ID
- qwen3-max-thinking
- Type
- Text generation
- Endpoints
- None has passed a live request yet
- Input
- text
- Output
- text
- Vision input
- Not supported
- Image editing
- Not supported
- Video input
- Not supported
- Reasoning
- Supported
- Streaming
- Supported
- Context window
- 256,000 tokens
Parameters by endpoint
The request field of each parameter on each endpoint, and the values it takes.
This model takes the fields in its sample request. It has no other stated parameters.
Pricing history
See how much this model saves and when its recorded price has been lowest.
Time zone
qwen3-max-thinking
Early dataSavings are visible now, but a best-time recommendation needs at least 12 observations across 6 different hours.
Last 24 hours
Average saved
35.0%
Last 24 hours
Cheapest time
Collecting data
Too early
Hourly coverage
3/24
3 total observations
Signal quality
Early
More history needed
Typical unit savings$1.05 / 1M input tokens$5.25 / 1M output tokens
When this model is cheapest
Each bar is the average discount for that hour in UTC (UTC). Missing bars mean no observation was recorded.
ObservedCheapestNo data
Hover a bar for savings per token and observation counts.
Source buckets are UTC; labels are converted to UTC.
Last 24 hours
—
of 424 ranked models
Last 7 days
—
of 440 ranked models
Last 30 days
—
of 440 ranked models
Ready when you are
Stop paying list price for tokens.
Marketplace pricing, capped at list, optimized further.