deepseek-v4-flash
- Input
- 0.15×
- Output
- 0.6×
- Cache read
- 0.003×
Models
Choose your intelligence.
Compare input, output and cache multipliers for the models in each access group. Select Details to check context tiers and time-based rules before connecting.
1× = 1 credit per 1,000 tokens. Input, output and cache are counted separately.
Zable AI
Consumption multiplier 1×deepseek-v4-pro
- Input
- 0.66×
- Output
- 1.98×
- Cache read
- 0.022×
glm-5.3
- Input
- 1.4×
- Output
- 4.4×
- Cache read
- 0.26×
glm-5.3-flash
- Input
- 0.15×
- Output
- 0.5×
- Cache read
- 0.03×
glm-5.3-flashx
- Input
- 0.37×
- Output
- 1.25×
- Cache read
- 0.075×
kimi-k2.6
- Input
- 0.95×
- Output
- 4×
- Cache read
- 0.16×
kimi-k2.7-code
- Input
- 0.95×
- Output
- 4×
- Cache read
- 0.19×
kimi-k3
- Input
- 3×
- Output
- 15×
- Cache read
- 0.3×
mimo-v2.5
- Input
- 0.14×
- Output
- 0.28×
- Cache read
- 0.0028×
mimo-v2.5-pro
- Input
- 0.435×
- Output
- 0.87×
- Cache read
- 0.0036×
minimax-m3
- Input
- 0.3×
- Output
- 1.2×
- Cache read
- 0.06×
qwen3.8-flash
- Input
- 0.113×
- Output
- 0.382×
- Cache read
- 0.014×
qwen3.8-max
- Input
- 1.65×
- Output
- 4.951×
- Cache read
- 0.206×
Standard rates for the selected access group. Context tiers and time-based rates may apply.