Qwen 3 Coder 480 B A35 B Instruct Fp 8 pricing
Source: litellm · updated 2026-10-27
Short answer: Qwen 3 Coder 480 B A35 B Instruct Fp 8 costs $0.450 per million input tokens and $1.80 per million output tokens. That is a 3:1 blended rate of $0.787 per million tokens, ranking it #359 of 704 tracked models by input price.
Price breakdown
| Metric | USD |
|---|---|
| Input, per 1M tokens | $0.450 |
| Output, per 1M tokens | $1.80 |
| Output รท input multiple | 4.00× |
| 3:1 blended, per 1M tokens | $0.787 |
| Context window | 262,144 tokens |
What a typical workload costs
| Monthly volume | Input cost | Output cost | Total | 1M in / 250K out | $0.450 | $0.450 | $0.900 |
|---|---|---|---|---|---|---|---|
| 10M in / 2.5M out | $4.50 | $4.50 | $9.00 | ||||
| 100M in / 25M out | $45.00 | $45.00 | $90.00 |
Other tensormesh models
| Model | Input | Output |
|---|---|---|
| Qwen 3.6 27 B Fp 8 | $0.320 | $3.20 |
| Qwen 3.5 397 B A17 B Fp 8 | $0.600 | $3.60 |
| GLM 5.1 Nvfp 4 Mtp | $1.40 | $4.40 |
Compare against
- Qwen 3 Coder 480 B A35 B Instruct Fp 8 vs Mimo V2.5 Pro
- Qwen 3 Coder 480 B A35 B Instruct Fp 8 vs Qwen 3 235 B A22 B Thinking 2507
Cheaper alternatives
Next cheaper option: Llama 3.3 70 B Instruct Fp 8 Fast at $0.293 per 1M input tokens. Next step up: Mistral Saba 24 B at $0.790 per 1M input tokens.