涵蓋 459 類工作流程,透過流量估算、產品功能與替代方案協助選擇。
“Inference provider”·383 個結果
重新開始一組相容 OpenAI 的 API,串接 500+ 圖像、影片、音訊與大模型


由 DeepSeek-V4 與 R1 驅動的高效開源權重與託管推理 AI,token 成本極低。

一個 API 串接 1500+ 可上線的生成式媒體模型

Together AI is the AI Native Cloud — a full-stack platform for running, fine-tuning and training open-source models. It spans serverless inference APIs, dedicated endpoints, GPU clusters, post-training and storage, all backed by in-house systems research.
Together AI is the AI Native Cloud — a full-stack platform for running, fine-tuning and training open-source models. It spans serverless inference APIs, dedicated endpoints, GPU clusters, post-training and storage, all backed by in-house systems research.

為代理快速提供開源模型推論,搭配熱快取路由


智譜 AI 的 GLM 大模型開放平台,多模態 API 與代理開發一次到位

Runware is an AI inference platform that exposes an enormous model catalog — images, video, audio, LLMs and 3D — through a single API endpoint, alongside serverless GPU compute and reserved capacity running on modular data centers the company designs and operates itself. It is aimed at developers and product teams who want to ship generative features without managing infrastructure.
Runware is an AI inference platform that exposes an enormous model catalog — images, video, audio, LLMs and 3D — through a single API endpoint, alongside serverless GPU compute and reserved capacity running on modular data centers the company designs and operates itself. It is aimed at developers and product teams who want to ship generative features without managing infrastructure.
Nebius is an AI-focused cloud platform that gives teams non-virtualized NVIDIA GPU clusters, managed MLOps tooling, and a serverless inference layer so they can move from experiment to production on one stack.
General Compute is an inference cloud built around alternative AI silicon rather than standard GPU fleets. The company buys racks of accelerators, racks them in colocation data centres, and handles the cooling, power, and networking itself. On top of that hardware it runs a disaggregated inference stack that splits each request: prompt processing runs on NVIDIA GPUs for prefill, while token generation runs on faster decode chips such as Cerebras, SambaNova, Positron AI, and d-Matrix. Customers get bare metal access or a tokens-per-second SLA.
General Compute is an inference cloud built around alternative AI silicon rather than standard GPU fleets. The company buys racks of accelerators, racks them in colocation data centres, and handles the cooling, power, and networking itself. On top of that hardware it runs a disaggregated inference stack that splits each request: prompt processing runs on NVIDIA GPUs for prefill, while token generation runs on faster decode chips such as Cerebras, SambaNova, Positron AI, and d-Matrix. Customers get bare metal access or a tokens-per-second SLA.
Unsloth is an open-source toolkit for running and training large language models locally. It provides faster, more memory-efficient fine-tuning, a local chat and agent environment with browser use, image and video generation, and an OpenAI-compatible API your existing tools can connect to.
Unsloth is an open-source toolkit for running and training large language models locally. It provides faster, more memory-efficient fine-tuning, a local chat and agent environment with browser use, image and video generation, and an OpenAI-compatible API your existing tools can connect to.

TypeSafe AI is an AI lab building machine-native intelligence infrastructure, and Jev is its first System One Model — a model designed to make decisions inside software rather than reply to humans. Instead of returning strings, Jev accepts structured questions and returns typed decisions with calibrated confidence scores, so applications can act autonomously when confidence is high and escalate for human review when it is not. TypeSafe reports roughly 193x faster and 444x cheaper performance on System One workloads compared with general-purpose LLMs.
TypeSafe AI is an AI lab building machine-native intelligence infrastructure, and Jev is its first System One Model — a model designed to make decisions inside software rather than reply to humans. Instead of returning strings, Jev accepts structured questions and returns typed decisions with calibrated confidence scores, so applications can act autonomously when confidence is high and escalate for human review when it is not. TypeSafe reports roughly 193x faster and 444x cheaper performance on System One workloads compared with general-purpose LLMs.


Relace is an inference provider serving open-weight frontier models through a per-token API, known for aggressive pricing and top-tier throughput on OpenRouter.

FluidStack builds and operates large-scale compute infrastructure for AI, including gigawatt-class data centers and GPU clusters used for training and inference. The company designs its own power, cooling and data center systems, onshores manufacturing and deploys solar generation to supply the power its clusters need. Rather than a self-serve cloud console, FluidStack works with AI labs and enterprises to provision large reserved clusters, with orchestration tooling for the ML frameworks and schedulers those teams already use.
FluidStack builds and operates large-scale compute infrastructure for AI, including gigawatt-class data centers and GPU clusters used for training and inference. The company designs its own power, cooling and data center systems, onshores manufacturing and deploys solar generation to supply the power its clusters need. Rather than a self-serve cloud console, FluidStack works with AI labs and enterprises to provision large reserved clusters, with orchestration tooling for the ML frameworks and schedulers those teams already use.



高吞吐開源權重模型推論,依延遲分級定價