Alternatives to Fireworks AI

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

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

Baseten serves custom and open-source models on dedicated GPUs with per-minute billing and token-priced Model APIs.

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.

Open-source LLM fine-tuning framework for LoRA, MoE, GRPO, and NVFP4 post-training on your own GPUs

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

PEFT is Hugging Face's free library for parameter-efficient fine-tuning of LLMs, vision, and diffusion models on consumer GPUs.

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

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

European GPU cloud with self-service GB300, B300 and B200 instances plus per-token open-weight inference

Crusoe Cloud GPU instances, serverless inference, and fine-tuning priced for AI teams that want orchestration included

Enterprise LLM fine-tuning with accuracy SLAs and sub-second inference.

Few-shot LLM fine-tuning platform that keeps models updated as new data arrives — private beta, access by email.

Modal's serverless GPU cloud runs Python functions and sandboxes with per-second billing, spinning up GPU inference and agent compute from

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

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

Self-hosted AI infrastructure platform turning your own GPUs into OpenAI- and Anthropic-compatible model endpoints

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

One API key for 1,000+ image, video, audio, 3D and LLM models on pay-per-use inference pricing

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

fal.ai runs 1,500+ generative media models — image, video, audio, 3D — through one API with per-output pricing.

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.

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

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

EnCharge AI builds analog in-memory computing silicon that runs AI inference at roughly 200 TOPS on about 8 watts.