Models

All models accessible through one API key. OpenAI-compatible endpoint.

DeepSeek V4 Flash

DeepSeek

89% cheaper
Input
$0.27/1M tokens
Output
$1.1/1M tokens
Context
128K
Speed
80 t/s
MMLU-Pro: 87.2vs GPT-4o $2.50/$10.00

Qwen 3.6 27B

Alibaba

92% cheaper
Input
$0.2/1M tokens
Output
$0.4/1M tokens
Context
128K
Speed
120 t/s
MMLU-Pro: 85.0vs GPT-4o $2.50/$10.00

GLM 5.2

Zhipu

94% cheaper
Input
$0.14/1M tokens
Output
$0.14/1M tokens
Context
128K
Speed
90 t/s
MMLU-Pro: 80.1vs GPT-4o $2.50/$10.00

Doubao Pro 256K

ByteDance

96% cheaper
Input
$0.1/1M tokens
Output
$0.3/1M tokens
Context
256K
Speed
100 t/s
MMLU-Pro: 82.5vs GPT-4o $2.50/$10.00