LLM API pricing by provider
- Cheapest: DeepSeek V4.1 Flash
- $0.0112
- Median
- $1.70
- Most expensive: o1-pro
- $600.00
Output price per million tokens · read October 20
Cheapest output, $ per million tokens
- DeepSeek V4.1 Flash $0.0112
- Mistral Nemo $0.03
- Gemma 3 4B $0.05
- S Llama 3 8B Lunaris $0.05
- Ling 3.0 Flash $0.063
- Mistral Small 3 $0.08
- Llama 3.1 8B Instruct $0.08
- gpt-oss-20b $0.09
As of October 20, 2026, 344 paid models in our catalog charge from $0.0112 to $600.00 per million output tokens, with a median of $1.80. Prices are list prices as published on OpenRouter.
A low price per token is not a low price per answer. Each provider page shows the price next to our coding index and, where a run cost exists, what a solved task costs.
Output price by maker
Output price per million tokens, 344 models. Lower is cheaper.
Each dot is one model's output price per million tokens. The log scale is needed: prices run from $0.0112 to $600.00. 5 models list a price of $0, which a log axis cannot show, so they are not drawn.
- OpenAI65
- Qwen42
- Google28
- Mistral AI20
- Z.ai16
- Anthropic15
- DeepSeek14
- Meta8
- Meta Llama8
- MiniMax8
- 44 other makers120
The widest spread is inside one maker: OpenAI lists models from $0.09 to $600.00. The dark tick on each row is that maker's median, so a maker's headline price and its typical price are two different numbers.
| Maker | Models plotted | Cheapest | Median | Dearest |
|---|---|---|---|---|
| OpenAI | 65 | $0.09 | $10.00 | $600.00 |
| Qwen | 42 | $0.13 | $1.00 | $12.00 |
| 28 | $0.05 | $3.00 | $12.00 | |
| Mistral AI | 20 | $0.03 | $0.90 | $7.50 |
| Z.ai | 16 | $0.16 | $1.98 | $8.80 |
| Anthropic | 15 | $0.50 | $25.00 | $75.00 |
| DeepSeek | 14 | $0.0112 | $1.00 | $2.15 |
| Meta | 8 | $0.20 | $4.25 | $20.00 |
| Meta Llama | 8 | $0.08 | $0.32 | $0.6525 |
| MiniMax | 8 | $0.95 | $1.20 | $2.20 |
| 44 other makers | 120 | $0.05 | $1.20 | $30.00 |
Source: OpenRouter. Fetched October 20, 2026.
Every provider's price range
The cheapest paid output price per million tokens, up to the dearest, and the provider's best model in our coding ranking. Each provider opens its full price list.