NEWKimi K3.1 Pro Abliterated (40% OFF!)

    Models

    Abliterated open-weight models on one OpenAI- and Anthropic-compatible endpoint. Context, vision, price, and speed below; open a model for its benchmarks and the one-line switch. Prices are per million tokens.

    ModelContextModalitiesInputOutputSpeedDetails
    GLM 5.3 Abliteratedglm-5.3-abliterated1,048,576$1.00$2.0075 tok/s
    GLM 5.3 Flash Abliteratedglm-5.3-flash-abliterated1,048,576$0.12$0.5080 tok/s
    DeepSeek V4.1 Flash 1017 Abliterateddeepseek-v4.1-flash-abliterated1,048,576$0.12$0.4580 tok/s
    DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-fast1,048,576$0.01$0.08100 tok/s
    Minimax M3.1 Flashminimax-m3.1-flash1,048,576$0.02$0.08—
    Kimi K3.1 Prokimi-k3.1-pro1,048,576$0.25$2.5048 tok/s
    DeepSeek V4.1 Flash 1017 Fast Abliterateddeepseek-v4.1-flash-abliterated-fast1,048,576$0.12$0.45200 tok/s
    Kimi K3.1 Pro Abliteratedkimi-k3.1-pro-abliterated1,048,576$1.50$3.0065 tok/s

    Full per-token rates, including the cache column, are on the pricing page.