Qwen: Qwen3 235B A22B vs Qwen3 Coder 480B A35B Instruct
Qwen3 Coder 480B A35B Instruct is cheaper than Qwen: Qwen3 235B A22B by about 100% on a blended (75% input / 25% output) basis. Qwen: Qwen3 235B A22B charges $0.230000 / $2.300000 per 1M input/output tokens; Qwen3 Coder 480B A35B Instruct charges $0.000000 / $0.000000.
| Field | Qwen: Qwen3 235B A22B | Qwen3 Coder 480B A35B Instruct |
|---|---|---|
| Input $/Mtok | $0.230000 | $0.000000 |
| Output $/Mtok | $2.300000 | $0.000000 |
| Cache read $/Mtok | – | – |
| Cache write $/Mtok | – | – |
| Input $/Mchar | 0.069354 | 0.000000 |
| Output $/Mchar | 0.693544 | 0.000000 |
| Context window | 128000 | 262144 |
| Tokenizer | qwen2 | qwen2 |
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 235B A22B | Qwen3 Coder 480B A35B Instruct |
|---|---|---|
| Coding agent | $3.33 | $0.00 |
| RAG chat | $5.40 | $0.00 |
| Summarisation (batch) | $4.37 | $0.00 |
| Classification | $2.71 | $0.00 |
| Chat assistant | $10.58 | $0.00 |
| Content generation | $18.86 | $0.00 |
Cite this page
Qwen: Qwen3 235B A22B vs Qwen3 Coder 480B A35B Instruct: $0.230000/$2.300000 vs $0.000000/$0.000000 per 1M tokens, as of 2026-10-04 (Token Barometer, https://tokenbarometer.com/compare/alibaba_qwen3-235b-a22b-vs-alibaba_qwen3-coder-480b-a35b).
Qwen: Qwen3 235B A22B page | Qwen3 Coder 480B A35B Instruct page | Open in calculator