Built by the team that solved sub-second cold starts for large language models.View benchmark

    Inference infrastructure for production AI.

    True serverless inferenceSub-second cold startsOpenAI-compatible APIs

    Production-ready Endpoints for open models and Sovereign Endpoints™ powered by the InferX runtime.

    Start BuildingExplore Models

    A request enters the OpenAI-compatible API, the scheduler restores initialized model state, GPU capacity attaches while tokens stream, and resources return to the pool when serving completes.

    INFERX RUNTIMEOPENAI-COMPATIBLE API
    SNAPSHOT READY
    RESOURCE LAYERCAPACITY IDLE
    Snapshot PoolInitialized model state
    GPU PoolGPU slicing enabled0 SLICES
    Secure RuntimeIsolated container
    Scale to ZeroCapacity released

    Compatible with your favorite AI tools

    OpenAILangChainOpenCodeHermesOpenClawContinueLlamaIndexVercel AI SDKDify
    Compatibility references do not imply partnership or endorsement.

    PROVEN ON PRODUCTION HARDWARE

    MODEL
    Qwen3.6-35B-A3B
    PRECISION
    FP16
    MODEL SIZE
    ~70 GB
    HARDWARE
    NVIDIA H100
    TRADITIONAL STARTUPLoad Model Weights≈35s
    INFERX STARTUPGPU Snapshot Restore<1s

    HOW THE STARTUP PATH CHANGES

    TRADITIONAL STARTUP

    • Request
    • Reserve GPU
    • Load model weights
    • Initialize runtime
    • Serve
    • GPU stays allocated

    INFERX STARTUP

    • Request
    • Scheduler
    • GPU Snapshot Restore
    • Allocate GPU resources
    • Serve
    • Scale to zero

    MONTHLY ACCESS

    An easy way to try InferX.

    Beta offer: $10/month, includes $50 in usage credits.

    MONTHLY ACCESS$20 / month $10 / month
    INCLUDED USAGE CREDITS$50 every billing cycle
    AFTER INCLUDED CREDITSPay per usage
    Start for $10

    Deployment options

    Choose how to build on InferX.

    The same runtime. Two deployment models.

    Endpoints

    Serverless

    Run frontier open models instantly.

    OpenAI-compatible APIs for production AI workloads with pay-per-token pricing.

    • Pay per token
    • OpenAI compatible
    • Frontier open models
    • Streaming responses
    • Long context
    • Tool calling

    Ideal for

    Applications, agents, prototypes, and production services.
    Explore Endpoints

    Sovereign Endpoints™

    Dedicated

    Your infrastructure. Your performance.

    Dedicated inference infrastructure powered by the InferX runtime.

    • Dedicated GPUs
    • Sub-second cold starts
    • Private networking
    • Predictable latency
    • Kubernetes
    • On-prem deployment

    Ideal for

    Enterprise production, regulated workloads, and private deployments.
    Deploy Endpoint
    Cold-start methodology
    InferX Runtime

    Inside the InferX Runtime

    A model-ready runtime, restored on demand.

    Explore the complete architecture
    1. 01Request
    2. 02Scheduler
    3. 03GPU Snapshot Restore
    4. 04Allocate GPU Resources
    5. 05Stream Tokens
    6. 06Scale to Zero
    7. 07Runtime Snapshot Ready

    OpenAI compatible

    Change one line. Keep your OpenAI code.

    Keep the OpenAI Python SDK and request shape. Point your existing client at InferX to serve open models in production.

    Verified SDK compatibility
    Get API KeyRead Docs
    inference.py
    PYTHON 3.12
    import osfrom openai import OpenAIclient = OpenAI(    api_key=os.environ["OPENAI_API_KEY"],    base_url="https://model.inferx.net/v1",)stream = client.chat.completions.create(    model="Qwen/Qwen3.6-35B-A3B",    messages=[{"role": "user", "content": "Hello, InferX."}],    stream=True,)for chunk in stream:    print(chunk.choices[0].delta.content or "", end="")

    Model platform

    200+ open models.
    Deploy any. Run popular models instantly.

    Access production-ready endpoints for popular models, or deploy any supported model as a dedicated endpoint with a single click.

    View published model catalog
    • Popular models ready instantly
    • Deploy any supported model
    • Pay per token or dedicated
    • OpenAI-compatible API
    • Powered by the InferX Runtime
    POPULAR + SUPPORTED MODELS
    Ready now

    DeepSeek V4

    Powered by the InferX Runtime

    General reasoningUse endpoint
    Ready now

    DeepSeek V4 Flash

    Powered by the InferX Runtime

    Low-latency reasoningUse endpoint
    Ready now

    Kimi K2

    Powered by the InferX Runtime

    Agentic workflowsUse endpoint
    Ready now

    Mimi

    Powered by the InferX Runtime

    Audio intelligenceUse endpoint
    Deployable

    Hy3

    Powered by the InferX Runtime

    Large-scale reasoningDeploy model
    Deployable

    Qwen3.6

    Powered by the InferX Runtime

    Production agentsDeploy model
    Deployable

    Gemma 4

    Powered by the InferX Runtime

    Multimodal workloadsDeploy model
    Deployable

    Llama

    Powered by the InferX Runtime

    General-purpose AIDeploy model
    Deployable

    Mistral

    Powered by the InferX Runtime

    Efficient inferenceDeploy model
    Deployable

    Phi

    Powered by the InferX Runtime

    Compact reasoningDeploy model
    Deployable

    GLM

    Powered by the InferX Runtime

    Multilingual agentsDeploy model
    Model 1 of 11Scroll or use arrow controls
    Explore unpublished models — ready to deploy with one click

    Two ways to build

    Choose how you run InferX.

    Consume inference01

    Endpoints

    Production inference managed by InferX.

    Start Building
    Run the runtime02

    Deploy InferX On-Prem

    Deploy the InferX Platform into Kubernetes, bare metal, private cloud, GPU clouds, or air-gapped environments.

    Explore On-Prem Deployment
    Want to understand how the platform works under the hood?Explore InferX Architecture

    Put production inference on a runtime built for it.

    Start BuildingContact Sales