Qwen4-Max API pricing
Official Qwen list price, last checked Nov 15, 2026. Updated Nov 15, 2026.
Qwen4-Max costs $1.50 per 1M input tokens and $4.50 per 1M output tokens, cached input at $0.075, and a 1M-token context window, per Qwen's official pricing page (checked Nov 15, 2026). It's a premium-priced model: number 59 of 86 by blended price ($2.25 per 1M tokens at a 3 to 1 input to output mix). On LMArena's Code Arena it ranks #6 of 54 models we price (relative score 72 out of 100). Nothing cheaper scores close to it on that board; DeepSeek V4.1 Flash is the nearest cheaper option.
Qwen4-Max price per million tokens
| Qwen4-Max price per million tokens | USD |
|---|---|
| Input | $1.50 |
| Cached input | $0.075 |
| Output | $4.50 |
| Blended, 3 parts input to 1 part output | $2.25 |
| Context window | 1M tokens |
Official note from Qwen: International (Singapore), launched Nov 16, 2026. List price for all prompts up to 1M tokens. Cache read $0.075; 5-minute cache write $1.875, 1-hour $3. Closed weights.
What Qwen4-Max costs for real work
List-price estimates for a month, no discounts. Our math, not a quote: change the token counts in the plan vs API calculator.
| Workload | Tokens a month | Qwen4-Max | GLM-5.3 | DeepSeek V4.1 Flash |
|---|---|---|---|---|
| Light: a side project or personal assistant | 2M in, 0.5M out | $5.25 | $5.00 | $1.20 |
| Medium: a small team or one product feature | 20M in, 5M out | $52.50 | $50.00 | $12.00 |
| Heavy: a production app | 200M in, 50M out | $525 | $500 | $120 |
| Medium with 80% of input cached | 20M in (16M cached), 5M out | $29.70 | $31.76 | $7.30 |
How Qwen4-Max ranks on LMArena
Relative score: 100 is the top model we price on that board. Rank is among the models in our price table, not the full leaderboard.
| Board | Rank | Score /100 |
|---|---|---|
| Coding (Code Arena) | #6 of 54 | 72 |
| Cheap bulk and agents (Agent Arena) | #11 of 42 | 40 |
Closest rivals to Qwen4-Max
The models nearest in price, other makers first. Ranks are on Code Arena, where we have them.
| Model | In / out | Blended | Verdict |
|---|---|---|---|
| GLM-5.3 | $1.40 / $4.40 | $2.15 | GLM-5.3 for price, Qwen4-Max for quality. Compare |
| GLM-5.2 | $1.40 / $4.40 | $2.15 | GLM-5.2 for price, Qwen4-Max for quality. Compare |
| GLM-5.1 | $1.40 / $4.40 | $2.15 | GLM-5.1 for price, Qwen4-Max for quality. Compare |
| GPT-6.1 Terra | $1 / $5 | $2 | GPT-6.1 Terra for price, Qwen4-Max for quality. Compare |
Cheaper alternatives to Qwen4-Max
At least 40% cheaper blended, best LMArena score first.
| Model | In / out | Blended | Verdict |
|---|---|---|---|
| DeepSeek V4.1 Flash | $0.30 / $1.20 | $0.525 | DeepSeek V4.1 Flash for price, Qwen4-Max for quality. Compare |
| MiMo-V2.6 Flash | $0.14 / $0.28 | $0.175 | MiMo-V2.6 Flash for price, Qwen4-Max for quality. Compare |
| Hunyuan Hy4 Preview | $0.834 / $2.50 | $1.25 | Hunyuan Hy4 Preview for price, Qwen4-Max for quality. Compare |
More from Qwen
- Qwen3.8 Max$2 in, $6 out per 1M
- Qwen3.8 2.4T-A95B$2 in, $6 out per 1M
- Qwen3.6 27B$0.60 in, $3.60 out per 1M
- Qwen3.5 397B-A17B$0.60 in, $3.60 out per 1M
- Qwen3.8 27B$0.50 in, $3 out per 1M
FAQ
How much does the Qwen4-Max API cost per million tokens?
$1.50 per 1M input tokens and $4.50 per 1M output tokens, plus $0.075 per 1M for cached input, as of Nov 15, 2026. A typical 3 to 1 input to output mix works out to $2.25 per 1M tokens.
Does Qwen4-Max charge more for long prompts?
Qwen doesn't list a long-prompt surcharge for Qwen4-Max. Its pricing note says: International (Singapore), launched Nov 16, 2026. List price for all prompts up to 1M tokens. Cache read $0.075; 5-minute cache write $1.875, 1-hour $3. Closed weights.
Is Qwen4-Max cheaper than GLM-5.3?
No. A medium month (20M input and 5M output tokens) costs $52.50 on Qwen4-Max and $50.00 on GLM-5.3 at list price. On Code Arena, Qwen4-Max ranks #6 and GLM-5.3 #24.
How does Qwen4-Max rank on LMArena?
On Code Arena it is #6 of 54 models we price, with a relative score of 72 out of 100. It ranks on 2 of 8 boards we track.
What is a cheaper alternative to Qwen4-Max?
DeepSeek V4.1 Flash costs $0.30 in and $1.20 out per 1M tokens, about 4.3x less blended, and scores 39 on Code Arena against 72 for Qwen4-Max.
Where does this Qwen4-Max price come from?
From Qwen's official pricing page, which we check every day. This page was last checked Nov 15, 2026.