GLM 5.1 API

    GLM 5.1 API.

    $0.90/M input, $3.00/M output (FP8). OpenAI-compatible, no contracts, no minimums.

    Get Started

    Point your OpenAI SDK at api.getlilac.com/v1 and request zai-org/glm-5.1.

    Model pricing

    Pay per token. No commitments.

    0.58s TTFT on shared warm endpoints. Competitive with aggregator listings at a lower per-token price.

    Model
    Context
    Input
    Output
    Latency
    Deprecates Apr 30Kimi K2.5

    Live now

    INT4
    262K
    $0.40/MCache $0.10/M
    $2.00/M
    0.38s TTFT
    Coming Apr 30Kimi K2.6INT4
    262K
    TBD
    TBD
    Coming Apr 30
    GLM 5.1

    Live now

    FP8
    203K
    $0.90/MCache $0.27/M
    $3.00/M
    0.58s TTFT
    Gemma 4 (31B)

    Live now

    BF16
    262K
    $0.11/M
    $0.35/M
    0.72s TTFT
    OpenAI-compatibleShared warm endpointsNo contractsNo minimums

    25% off all tokens above 1B/month for 3 months. That is $0.30/M input and $1.50/M output above the threshold.

    More models are coming soon and will be added as they go live.

    Integration

    One base URL change.

    Keep the OpenAI SDK and point it at Lilac. Your existing code just works.

    inference.py

    from openai import OpenAI

    client = OpenAI(

    base_url="https://api.openai.com/v1",

    api_key="sk_...",

    )

    response = client.chat.completions.create(

    model="openai/gpt-5.4",

    messages=[{"role": "user", "content": "Hello!"}],

    )

    # Same code. Same SDK. Fraction of the price.

    01

    Standard OpenAI client — just change the base URL.

    02

    Visible pricing. No aggregator markup.

    03

    Shared warm endpoints, no cold starts.

    Frequently asked questions

    How do I call the API?

    Set base_url to https://api.getlilac.com/v1 in the OpenAI SDK, model name zai-org/glm-5.1.

    How much does it cost?

    $0.90/M input, $3.00/M output on the shared endpoint.

    Is my data stored or used for training?

    No. Lilac is ZDR (Zero Data Retention) compliant — prompt and completion data is never stored or used for model training.

    Start running inference in minutes.

    No contracts, no commitments. Swap your base URL and pay less for the same output quality.

    Get Started

    No commitment required.