BiWu.AiCheapest LLM API: Models Ranked by Price per 1M Tokens
We rank 36 official LLM API models from cheapest to most expensive. The cheapest right now is DeepSeek V4 Flash at $0.14 per 1M input tokens and $0.28 per 1M output tokens.
The ranking uses a blended price that assumes 3 input tokens for every output token, which is close to typical chat and coding use. Open the calculator to rank models for your own token mix.
Prices checked Oct 10, 2026
Models ranked
36
Cheapest model
DeepSeek V4 Flash
$0.14 / $0.28
Blended price
$0.18/1M
3:1 input to output
Prices checked
Oct 10, 2026
LLM API models ranked by price
Official prices per 1 million tokens. The blended price weights input and output 3:1.
| Rank | Model | Family | Input | Output | Blended (3:1) |
|---|---|---|---|---|---|
| 1 | DeepSeek V4 Flash | DeepSeek | $0.14 | $0.28 | $0.18 |
| 2 | MiMo V2.5 | MiMo | $0.14 | $0.28 | $0.18 |
| 3 | GPT-6 Luna | OpenAI | $0.10 | $0.50 | $0.20 |
| 4 | Qwen3.5 Plus | Qwen | CN¥0.80 (≈ $0.12) | CN¥4.80 (≈ $0.72) | $0.27 |
| 5 | Step 3.7 Flash | StepFun | CN¥1.35 (≈ $0.20) | CN¥8.10 (≈ $1.21) | $0.45 |
| 6 | GPT-5.4 Nano | OpenAI | $0.20 | $1.25 | $0.46 |
| 7 | Qwen3.7 Plus | Qwen | CN¥2 (≈ $0.30) | CN¥8 (≈ $1.19) | $0.52 |
| 8 | MiniMax M3 | MiniMax | $0.30 | $1.20 | $0.52 |
| 9 | MiniMax M2.7 | MiniMax | $0.30 | $1.20 | $0.52 |
| 10 | MiniMax M2.5 | MiniMax | $0.30 | $1.20 | $0.52 |
| 11 | DeepSeek V4 Pro | DeepSeek | $0.435 | $0.87 | $0.54 |
| 12 | MiMo V2.5 Pro | MiMo | $0.435 | $0.87 | $0.54 |
| 13 | Gemini 3.1 Flash-Lite | Gemini | $0.25 | $1.50 | $0.56 |
| 14 | Qwen3.6 Plus | Qwen | CN¥2 (≈ $0.30) | CN¥12 (≈ $1.79) | $0.67 |
| 15 | Qwen3 Coder Plus | Qwen | CN¥4 (≈ $0.60) | CN¥16 (≈ $2.39) | $1.04 |
| 16 | GLM-5 | GLM | CN¥4 (≈ $0.60) | CN¥18 (≈ $2.68) | $1.12 |
| 17 | Gemini 3 Flash Preview | Gemini | $0.50 | $3 | $1.13 |
| 18 | Kimi K2.5 | Kimi | CN¥4 (≈ $0.60) | CN¥21 (≈ $3.13) | $1.23 |
| 19 | GLM-5.1 | GLM | CN¥6 (≈ $0.89) | CN¥24 (≈ $3.58) | $1.57 |
| 20 | GPT-5.4 Mini | OpenAI | $0.75 | $4.50 | $1.69 |
| 21 | Kimi K2.6 | Kimi | CN¥6.50 (≈ $0.97) | CN¥27 (≈ $4.03) | $1.73 |
| 22 | Claude Haiku 4.5 | Claude | $1 | $5 | $2.00 |
| 23 | Qwen3.7 Max | Qwen | CN¥12 (≈ $1.79) | CN¥36 (≈ $5.37) | $2.68 |
| 24 | GPT-6 Sol | OpenAI | $2 | $10 | $4.00 |
| 25 | Claude Sonnet 5 | Claude | $2 | $10 | $4.00 |
| 26 | Gemini 3.1 Pro Preview | Gemini | $2 | $12 | $4.50 |
| 27 | GPT-5.4 | OpenAI | $2.50 | $15 | $5.63 |
| 28 | Kimi K3 | Kimi | CN¥20 (≈ $2.98) | CN¥100 (≈ $14.91) | $5.96 |
| 29 | Claude Sonnet 4.6 | Claude | $3 | $15 | $6.00 |
| 30 | Claude Opus 5 | Claude | $5 | $25 | $10.00 |
| 31 | Claude Opus 4.8 | Claude | $5 | $25 | $10.00 |
| 32 | Claude Opus 4.7 | Claude | $5 | $25 | $10.00 |
| 33 | Claude Opus 4.6 | Claude | $5 | $25 | $10.00 |
| 34 | GPT-5.5 | OpenAI | $5 | $30 | $11.25 |
| 35 | GPT-6 Astra | OpenAI | $10 | $50 | $20.00 |
| 36 | Claude Fable 5 | Claude | $10 | $50 | $20.00 |
Cheapest model from each provider
| Family | Model | Input | Output |
|---|---|---|---|
| DeepSeek | DeepSeek V4 Flash | $0.14 | $0.28 |
| MiMo | MiMo V2.5 | $0.14 | $0.28 |
| OpenAI | GPT-6 Luna | $0.10 | $0.50 |
| Qwen | Qwen3.5 Plus | CN¥0.80 (≈ $0.12) | CN¥4.80 (≈ $0.72) |
| StepFun | Step 3.7 Flash | CN¥1.35 (≈ $0.20) | CN¥8.10 (≈ $1.21) |
| MiniMax | MiniMax M3 | $0.30 | $1.20 |
| Gemini | Gemini 3.1 Flash-Lite | $0.25 | $1.50 |
| GLM | GLM-5 | CN¥4 (≈ $0.60) | CN¥18 (≈ $2.68) |
| Kimi | Kimi K2.5 | CN¥4 (≈ $0.60) | CN¥21 (≈ $3.13) |
| Claude | Claude Haiku 4.5 | $1 | $5 |
How this ranking works
- Only official list prices are used: the provider's own API, standard tier, short-context rate.
- Batch, flex, priority and long-context rates are not included and can be cheaper or more expensive.
- Cached input can lower the real cost a lot when the same prompt prefix is reused. The calculator lets you set a cache share.
- Free tiers and trial credits are not counted, because their limits change often.
FAQ
What is the cheapest LLM API?
DeepSeek V4 Flash (DeepSeek) is the cheapest of the 36 models we track: $0.14 per 1M input tokens and $0.28 per 1M output tokens at the official price. Prices checked Oct 10, 2026.
What is the cheapest OpenAI API model?
GPT-6 Luna is the cheapest OpenAI model on the official API: $0.10 per 1M input tokens and $0.50 per 1M output tokens.
What is the cheapest Claude API model?
Claude Haiku 4.5 is the cheapest Claude model on the official API: $1 per 1M input tokens and $5 per 1M output tokens.
What is the cheapest Gemini API model?
Gemini 3.1 Flash-Lite is the cheapest Gemini model on the official API: $0.25 per 1M input tokens and $1.50 per 1M output tokens.
What is the cheapest DeepSeek API model?
DeepSeek V4 Flash is the cheapest DeepSeek model on the official API: $0.14 per 1M input tokens and $0.28 per 1M output tokens.
Related pricing
- LLM API cost calculator
- Compare all LLM API prices
- OpenAI API pricing
- Claude API pricing
- Gemini API pricing
- DeepSeek API pricing
- Qwen API pricing
- Kimi API pricing
- MiniMax API pricing
- GLM API pricing
- Xiaomi MiMo API pricing
- Grok API pricing
- Mistral API pricing
- Claude Opus API pricing
- Claude Sonnet API pricing
- GPT-5 API pricing
- Gemini Pro API pricing
BiWu.Ai lists public API prices from official pricing pages for reference and does not sell API access. Prices can change without notice; check the provider's pricing page before you build on it.