Qwen 3.8 B pricing
Source: litellm · updated 2026-10-27
Short answer: Qwen 3.8 B costs $0.040 per million input tokens and $0.140 per million output tokens. That is a 3:1 blended rate of $0.065 per million tokens, ranking it #53 of 704 tracked models by input price.
Price breakdown
| Metric | USD |
|---|---|
| Input, per 1M tokens | $0.040 |
| Output, per 1M tokens | $0.140 |
| Output รท input multiple | 3.50× |
| 3:1 blended, per 1M tokens | $0.065 |
| Context window | 32,768 tokens |
What a typical workload costs
| Monthly volume | Input cost | Output cost | Total | 1M in / 250K out | $0.040 | $0.035 | $0.075 |
|---|---|---|---|---|---|---|---|
| 10M in / 2.5M out | $0.400 | $0.350 | $0.750 | ||||
| 100M in / 25M out | $4.00 | $3.50 | $7.50 |
Other llamagate models
| Model | Input | Output |
|---|---|---|
| Llama 3.1.8 B | $0.030 | $0.050 |
| Gemma 3.4 B | $0.030 | $0.080 |
| Llama 3.2.3 B | $0.040 | $0.080 |
| DeepSeek Coder 6.7 B | $0.060 | $0.120 |
| Codellama 7 B | $0.060 | $0.120 |
Compare against
Cheaper alternatives
Next cheaper option: Gemma 4 26 B at $0.030 per 1M input tokens. Next step up: Command R7 B at $0.037 per 1M input tokens.