Alternatives to Etched AI

Rebellions builds chiplet-based AI inference hardware — Rebel100 silicon, RebelCard, RebelServer, RebelRack and RebelPOD for enterprise LLM

Rack-scale AI inference system using log-space math to serve multi-trillion-parameter models at 608 PFLOPS per rack.

Groq delivers fast LLM inference on its own LPU and LPX silicon through an OpenAI-compatible API.

Cerebras inference cloud runs frontier LLMs in Ultrafast mode at up to 750 tokens/sec for real-time agents and voice AI.

MAX Engine delivers OpenAI-compatible inference of open models across NVIDIA, AMD, Apple silicon, CPUs, and ASICs from one package.

Fast open-weight LLM inference on SambaNova's own RDU silicon — MiniMax M3.1 Flash, MiniMax M2.7, DeepSeek-V3.1 and gpt-oss-120b.

One inference API for 400K+ image, video, audio, LLM and 3D models, plus serverless GPUs starting at $0.63/GPU-hour reserved.

Serverless AI inference on 100+ open models via an OpenAI-compatible API, billed per token with no GPU overhead.

TensorRT-LLM: NVIDIA's open-source library for high-throughput LLM and visual generation inference

No-code LLM fine-tuning GUI for private, air-gapped model customization — H2O LLM Studio delivers sovereign AI training.

Serverless GPU inference for custom ML models — deploy in minutes, billed per second with scale-to-zero.

Multimodal inference API on dedicated NVIDIA GPUs with an OpenAI-compatible endpoint

vLLM turns open-weight LLMs into high-throughput, OpenAI-compatible inference endpoints on your own GPUs.

Flat-rate subscription offering unlimited inference on open models GLM, Qwen, DeepSeek, and Kimi, tuned continuously for your traffic type.

Run Llama 3.1 405B and other huge language models at home over a BitTorrent-style decentralized inference network.

KubeAI is a Kubernetes operator for self-hosted LLM, embedding, reranking, and speech-to-text inference with an OpenAI-compatible API.

Mini Infer is a readable Python LLM inference engine with hand-written CUDA and Triton kernels for learning LLM serving.

Run fast, self-hosted inference for LLMs, vision-language and diffusion models with SGLang serving.

Together AI is an AI cloud for open-weight models: per-token inference APIs with GPU clusters underneath.

Managed Ray platform for distributed model training, multimodal data curation, and batch embedding generation

Rent GPU pods, Serverless inference endpoints and multi-node GPU clusters by the second on Runpod's AI developer cloud.

One API for image, video, speech, music and text generation with hosted MCP agents

Novita AI runs 200+ open-weight LLM, image, audio, video and Vision models through one OpenAI-compatible serverless API.

CoreWeave is the AI-native GPU cloud for training, reinforcement learning post-training, and low-latency inference on NVIDIA cluster-scale

Lilac turns idle GPU clusters into a white-labeled neocloud with spot H100 leases and serverless inference.

FuriosaAI's RNGD inference accelerators deliver LLM serving in 3 kW air-cooled racks behind an OpenAI-compatible API.

Parallax is a distributed model serving framework for building your own self-hosted AI cluster across scattered GPUs, Macs, and

Predibase's LoRA fine-tuning lineage now sits inside Rubrik's AI agent governance platform

Mesh LLM splits large open-weight models into layer ranges across your own GPUs and serves them from one OpenAI-compatible endpoint.

Open-source KV cache layer for LLM inference that skips redundant prefill work