software catalog

Updated Sep 1, 2026

Supported Features

← Docs index

What Intel® Inference Microservices supports today, and which capabilities it passes straight through to vLLM for you to configure yourself. Use this page to check an assumption before you design around it.

Everything below describes the published images at tag 0.1.0.

Validated capabilities

These are part of the product surface and covered by the built-in configurations.

CapabilityStatusWhere
POST /v1/chat/completionsSupportedAPI
POST /v1/completionsSupportedAPI
GET /v1/modelsSupported — returns one modelAPI
GET /healthSupported — 200 ready, 503 not readyAPI
GET /metrics (Prometheus)SupportedAPI
Streaming ("stream": true)SupportedAPI
OpenAI SDK, LangChain, LlamaIndex, LiteLLM, HaystackSupported — base URL onlyAPI
Automatic hardware detection and configurationSupportedMeet
Kubernetes Deployment + Service with probesSupportedDeploy on Kubernetes
Docker container health checkSupported — built into the imageQuickstart
Pre-downloaded model cacheSupportedModel caching
Pinning a built-in configurationSupported — INFERENCE_PROFILE_IDEnvironment variables
Non-root container, no shellSupportedSecurity
Model-provided code executionGranted only where the model requires it; INFERENCE_TRUST_REMOTE_CODE decides explicitlySecurity

vLLM pass-throughs

INFERENCE_ENGINE_ARGS merges a JSON object of vLLM flags over the built-in configuration. That makes the following reachable — but the resulting combination is yours, not the Intel®-validated one for that model, so verify it with dry-run and load-test it before you rely on it.

CapabilityHowCaveat
Tool / function calling{"enable-auto-tool-choice": true, "tool-call-parser": "<parser>"}The parser is model-specific and must match the model's template
Structured / JSON output{"guided-decoding-backend": "xgrammar"} plus per-request response_formatBackend and model support vary
Shorter or longer context{"max-model-len": 4096}Raising it above the tuned value increases KV-cache memory and can cause OOM
Prefix caching{"enable-prefix-caching": true}Trades memory for repeated-prompt latency
LoRA adapters{"enable-lora": true, "lora-modules": [...]}Adapters are your responsibility to mount and validate
Reasoning parsers{"reasoning-parser": "<parser>"}Model-specific

A bad value fails in one of two ways, and they look nothing alike:

What you passedWhat happens
Malformed JSON, or JSON that is not an objectIgnored with a warning. The container serves with the built-in configuration.
Valid JSON, but a flag vLLM rejectsThe container fails at startup. vLLM's own parser validates the flags before the engine launches.

dry-run --format json tells you which one you have. It either prints the resolved flags in "command_script" or fails with the same error serve would give.

Multimodal models

Two catalog entries are tagged multimodal: Gemma E4B Instruct and Qwen3-VL 30B-A3B Instruct. For those images, image inputs use the standard OpenAI content-parts shape that vLLM implements, on the same /v1/chat/completions endpoint — there is no separate endpoint and no Intel®-specific request format. Input types accepted per model follow the model card; see the vLLM multimodal inputs reference for the exact request shape.

Text-only catalog entries do not accept image parts.

Processors and platforms

Supported
ProcessorsIntel® Xeon® CPU (xeon): Intel® Xeon® 4, Intel® Xeon® 5, and Intel® Xeon® 6 processors (Sapphire Rapids, Emerald Rapids, Granite Rapids, Sierra Forest) — see Supported Intel® platforms
EnginevLLM (INFERENCE_ENGINE=vllm)
Accelerator familyIntel® Xeon® CPU (INFERENCE_ACCELERATOR_TYPE=xeon)
Precisionbf16 (auto resolves to the same)
OrchestrationKubernetes 1.24+ — Deployment plus Service
Single hostLinux + Docker 20.10+ (Ubuntu 22.04 LTS or later recommended)
Where you find image namesIntel® Software Catalog

Details and sizing: Supported Intel® platforms.

Scaling

One container serves one model, and GET /v1/models returns a single entry. To serve several models, run one Deployment per model and route between them in your application or with a gateway such as LiteLLM. To serve more traffic for one model, add replicas behind the Service. Keep the readiness probe on /health so traffic reaches a pod only after its engine has loaded the model. Watch vllm:num_requests_waiting to decide when to add replicas (API).

  • API — endpoints, streaming, tool calling, metrics
  • Environment variables — every setting, including INFERENCE_ENGINE_ARGS
  • FAQ — short answers to common questions
  • Glossary — profiles, NUMA, AMX, bf16, tensor parallelism