Providers

ccapi

@ccapi

CC Max & codex PRO, stable RPM.

AnyRouter

@anyrouter

Providing fast and stable AI models.

Gonka

@gonka

OpenAI-compatible inference gateway built on top of the decentralized AI network.

AI MAX

@aizhongzhuanzhan

Provide high-quality AI services.

Darek-Store

@darek

Một nhà bán API có mọi loại model gpt, gemini, claude

FastAPI

@fastapi

Provide stable AI services.

zrouter

@zrouter

Multi-model AI inference provider with OpenAI-compatible access to frontier and open models. Built for developers, agents, coding tools, automation, and AI applications.

云集AI

@altman

云集AI，聚合全球 AI 能力。 本团队长期深耕 AI 账号与资源服务行业，现正式进入上层聚合中转领域。核心账号自开自用、多平台自持，不依赖单一公共池，稳定高可用，支持高并发、弹性扩容与持续上量。 云集AI 通过统一 API 接口，聚合全球主流 AI 厂商与多模态模型能力。OpenAI 兼容，一个 Key 调用多家模型，接入简单，适合个人开发者、团队与企业项目长期使用。

shengapi

@shengapi

Gpt and claude

Pool AI extremely cheap

@pal

Sell original AI models at low prices, including major AI models such as GPT, Gemini, and Claude. We guarantee that the models are genuine—no substitution or fake models.

NexLink AI

@nexlink

NexLink AI is a one-stop AI API gateway for developers, teams, and enterprises. With an OpenAI-compatible interface, it brings leading global AI providers and multimodal models into a single endpoint. One key gives you access to multiple models, making switching and integration far simpler. Built on self-owned accounts across multiple platforms—not a single shared pool—NexLink AI supports high concurrency, elastic scaling, and sustained traffic growth, helping AI projects move from prototype to production faster.

Hyperfusion

@hyperfusion

Hyperfusion is a high-performance AI routing platform offering access to a curated selection of leading models through a unified, developer-friendly API. Built with sovereignty and efficiency in mind and proudly hosted in UAE

Singularityapi

@singularityapi

SingularityAPI is a unified AI inference gateway — a single OpenAI-compatible endpoint that aggregates multiple models from the world's leading AI labs. One API key, one base URL, and you get access to the full frontier: DeepSeek V3.2 and V4, GPT-5.6 Luna/Sol/Terra, Kimi K2.6/K3, and more models. We are the cheapest inference provider in the market.

VoidEZ

@voidez

VoidEZ provides cost-efficient OpenAI-compatible access to selected GPT-class models. We focus on stable streaming, competitive token pricing, low overhead, and controlled concurrency to deliver reliable inference for everyday and coding workloads.

Teddy_Node_01

@teddy-node-01

Teddy_Node_01 is an OpenAI-compatible inference gateway hosted in Helsinki, Finland. It exposes /v1/chat/completions, /v1/models, and /v1/probe over HTTPS with PM2 cluster mode, real-time health checks, and upstream fallback routing across OpenAI, Groq, and DeepSeek-compatible routes. Supported text models include openai/gpt-4o and meta-llama/llama-3-70b-instruct.

TokenHub

@tokenhub

Self-hosted open-weight LLM inference with an OpenAI-compatible API. Owned GPU capacity serving open models with transparent per-token pricing and steady low-latency responses for chat completions.

Thanos

@thanos

Provide high-quality AI services.

Engineering Home

@engineeringhub

Our company specializes in accelerating Large Language Models (LLMs) on NVIDIA GPUs ranging from consumer grade to entry level professional cards. To achieve high inference performance, we employ a set of algorithmic innovations, including: • DynamicSparse – an advanced approach that leverages NVIDIA’s hardware support for sparse computations. Unlike conventional sparse methods, our technique does not require additional fine tuning or model conversion. Sparsification is performed on the fly within custom CUDA kernels of our software stack, using the inherent activation sparsity present in LLMs, thereby preserving output quality. • Block wise Quantization for Communication – our software implements an original block quantization algorithm for data exchanged between multiple GPUs. This enables efficient multi GPU scaling without relying on NVLink, which is often absent in low end accelerators. • Diffusion MTP – we employ proprietary Multi Token Prediction (MTP) blocks based on diffusion models. These blocks significantly reduce memory footprint, which is especially crucial for accelerators that do not feature HBM memory. • CPU Offloading – our runtime automatically offloads certain operations and low activity experts in Mixture of Experts (MoE) architectures to the host CPU, thereby boosting overall server throughput. • Low Precision Formats – the software adaptively utilizes reduced precision data types, including NV/MXFP4, NF4, and others, to optimize data transfers and further increase performance. All of the above techniques are integrated as a custom compiler layer within the vLLM framework, enabling seamless execution of virtually any modern LLM. Our core mission is to develop cost effective inference solutions for smaller models on affordable hardware.

Mixlayer

@mixlayer

Mixlayer provides frontier-grade infrastructure for open source AI models.

Sin

@sin

Providing stable services

Tiyuvta

@tiyuvta

Independent Blackwell inference provider for Qwen3.6 with OpenAI-compatible streaming, tool calling, structured outputs, and transparent metering.

Transparent Compute

@tcast-ai

Dynamic elastic load serving provider for open source models. We provide consistant service at competitive pricing.