Qwen: Qwen3 Max vs Meta: Llama 4 Maverick
Meta: Llama 4 Maverick is cheaper than Qwen: Qwen3 Max by about 96% on a blended (75% input / 25% output) basis. Qwen: Qwen3 Max charges $0.780000 / $3.900000 per 1M input/output tokens; Meta: Llama 4 Maverick charges $0.050000 / $0.100000.
| Field | Qwen: Qwen3 Max | Meta: Llama 4 Maverick |
|---|---|---|
| Input $/Mtok | $0.780000 | $0.050000 |
| Output $/Mtok | $3.900000 | $0.100000 |
| Cache read $/Mtok | $0.156000 | – |
| Cache write $/Mtok | $0.975000 | – |
| Input $/Mchar | 0.235202 | 0.015047 |
| Output $/Mchar | 1.176009 | 0.030093 |
| Context window | 262144 | 1048576 |
| Tokenizer | qwen2 | llama3 |
Price history
Same workload, monthly cost
Illustrative monthly cost at 10M tokens/month, split per workload preset. Not a substitute for the calculator, which uses your own volume, cache and batch assumptions.
| Workload | Qwen: Qwen3 Max | Meta: Llama 4 Maverick |
|---|---|---|
| Coding agent | $9.36 | $0.53 |
| RAG chat | $12.48 | $0.58 |
| Summarisation (batch) | $10.92 | $0.55 |
| Classification | $8.42 | $0.51 |
| Chat assistant | $20.28 | $0.70 |
| Content generation | $32.76 | $0.90 |
Cite this page
Qwen: Qwen3 Max vs Meta: Llama 4 Maverick: $0.780000/$3.900000 vs $0.050000/$0.100000 per 1M tokens, as of 2026-10-04 (Token Barometer, https://tokenbarometer.com/compare/alibaba_qwen3-max-vs-meta_llama-4-maverick).
Qwen: Qwen3 Max page | Meta: Llama 4 Maverick page | Open in calculator