AnyscaleAI Model APIs & GPU CloudAnyscale is the company behind Ray. Its managed platform runs training, batch inference and LLM serving on autoscaling Ray clusters, hosted or BYOC.CommercialSelf-hostedPaidViews · 7 daysTop 29%
AI Model APIs & GPU Cloud
Hosted model APIs, inference endpoints and GPU clouds for running, fine-tuning and serving AI models.
15 results in AI Model APIs & GPU Cloud
Active Pro listings from this category appear first. Your sort applies within Pro and Free results.
AnyscaleAI Model APIs & GPU CloudAnyscale is the company behind Ray. Its managed platform runs training, batch inference and LLM serving on autoscaling Ray clusters, hosted or BYOC.CommercialSelf-hostedPaid23 views
BifrostLLM App Platforms & RAGBifrost is an Apache-2.0 AI gateway written in Go. It unifies 20+ providers behind one OpenAI-compatible API with fallbacks, virtual keys, semantic caching, logging and an MCP gateway, run through NPX or Docker.Open sourceSelf-hostedViews · 7 daysNo views in 7 days
CohereAI Model APIs & GPU CloudCohere offers Command language models, Embed and Rerank via API, plus the North platform, with free rate-limited trial keys and pay-as-you-go production keys.CommercialSelf-hostedFreemiumViews · 7 daysTop 29%
DeepInfraAI Model APIs & GPU CloudDeepInfra is an AI inference cloud that serves 200+ open and hosted models through OpenAI- and Anthropic-compatible APIs, billed per token or execution time, plus dedicated GPU deployments, GPU instances, sandboxes and batch jobs.CommercialWeb platformPaidViews · 7 daysTop 29%
ElevenLabsProAI Voice, Speech & TranscriptionElevenLabs is a closed-source voice AI platform: text to speech, Scribe speech to text, dubbing, music and ElevenAgents voice agents through a web app and API. Official SDKs are MIT licensed; the plans start free.CommercialWeb platformFreemiumViews · 7 daysTop 3%
Fireworks AIAI Model APIs & GPU CloudFireworks AI runs open models through pay-per-token serverless inference, reserved throughput and inference inside your own cloud, plus training and routing.CommercialSoftwarePaidViews · 7 daysNo views in 7 days
LLM GatewayLLM App Platforms & RAGAGPL-3.0 API gateway for LLMs: one OpenAI-compatible endpoint for many providers, with routing, failover, key management, usage and cost analytics, caching and guardrails. Use the managed cloud or self-host with Docker.Open sourceSelf-hostedFreemiumViews · 7 daysTop 29%
Mistral AIAI Model APIs & GPU CloudMistral AI offers open-weight and commercial language models through Studio and its API, plus Vibe, an agent for productivity and coding.Open sourceSelf-hostedFreemiumViews · 7 daysNo views in 7 days
ModalAI Model APIs & GPU CloudModal is a cloud platform for running AI inference, training and sandboxes on GPUs and CPUs billed per second, with $30 of free credits each month.CommercialSoftwareFreemiumViews · 7 daysTop 29%
OctoAIAI Model APIs & GPU CloudOctoAI was a hosted inference service for open-source text and image models, plus OctoStack for running the same serving stack in a customer's own environment. The company was acquired by NVIDIA in 2024 and its commercial services ended on October 31, 2024.CommercialWeb platformPaidViews · 7 daysTop 16%
OllamaLocal LLMs & InferenceOllama is an MIT-licensed tool for downloading and running open models such as Gemma, Qwen, DeepSeek and gpt-oss on your own machine. It ships a desktop app, CLI and local REST API, and offers cloud-hosted models on a pay-as-you-go or subscription basis.Open sourceSelf-hostedFreemiumViews · 7 daysTop 29%
ReplicateAI Model APIs & GPU CloudReplicate runs public and custom machine learning models through an API, billed by usage, with Cog as the open-source tool for packaging your own models.CommercialSoftwarePaidViews · 7 daysTop 29%
RunpodAI Model APIs & GPU CloudRunpod is a GPU cloud from Runpod Inc. offering container Pods on Secure or Community Cloud, autoscaling Serverless endpoints, multi-node Instant Clusters and hosted model APIs, with per-second billing, a REST API, runpodctl and SDKs.CommercialWeb platformPaidViews · 7 daysTop 3%
Together AIAI Model APIs & GPU CloudTogether AI provides serverless and dedicated inference, fine-tuning and post-training, and GPU clusters for teams running open-source models.CommercialSoftwarePaidViews · 7 daysNo views in 7 days
Views · 7 daysNo views in 7 days
Shortlists for AI Model APIs & GPU Cloud
Explore collections with tools relevant to this category, grouped by deployment, source model or workload.
IT tools with an API →
15 matching tools from this category
Self-hosted IT tools →
6 matching tools from this category
Open-source IT tools →
4 matching tools from this category
How to choose ai model apis & gpu cloud tools
Model APIs and GPU clouds let teams run AI models without owning accelerators: hosted endpoints answer requests billed per token or per second, while rented GPU instances and clusters handle custom inference, fine-tuning and training. Some providers serve their own proprietary models, others host open-weight models or let you package and deploy your own. This differs from local inference tools, which run models on hardware you already control.
Who uses these tools
- Application developers adding model calls to products
- ML engineers fine-tuning and serving open-weight models
- Startups needing burst GPU capacity without hardware purchases
Capabilities to compare
- Serverless inference endpoints with per-token billing
- Dedicated GPU instances and multi-node clusters
- Fine-tuning and custom model deployment
- Embeddings, reranking, speech and image models
- Compatibility with common chat completion API formats
Operating considerations
- Data retention and training use of submitted prompts
- Region availability for data residency requirements
- Cold starts and capacity limits on serverless endpoints
- Provider shutdowns forcing quick migration of workloads
Selection checklist
- Estimate monthly token or GPU hours from real traffic and price it on each shortlisted provider.
- Benchmark latency and output quality with your own prompts rather than relying on published leaderboards.
- Read the data processing terms to confirm prompts are not retained or used for training.
- Keep an abstraction layer or compatible API so you can switch providers if one changes terms or closes.
What belongs in this category: A provider fits this category when it sells remote access to model inference or to GPU compute for AI workloads, rather than shipping software you install to run models locally.
AI Model APIs & GPU Cloud on iThub in figures
iThub lists 15 tools in AI Model APIs & GPU Cloud. 4 of 15 are open source, 6 of 15 can be self-hosted, 7 of 15 have a free, freemium or open-source pricing model and 15 of 15 offer an API.
- Tools listed
- 15
- Open source
- 4
- Self-hostable
- 6
- Free to start
- 7
- With an API
- 15
- Web app
- 7
- Desktop app
- 1
Typical use cases
- Serving an open-weight model behind a product feature
- Fine-tuning a model on domain data for classification
- Running batch inference jobs over large document sets
- Renting GPUs for short training or evaluation experiments
Trade-offs to model
- Serverless endpoints need no capacity planning but cost more per request at scale than reserved or dedicated GPUs.
- Proprietary model APIs often lead on quality, while hosted open-weight models allow portability and self-hosting later.
Common mistakes
- Building directly against one provider's proprietary SDK, which makes a later migration slow and expensive.
- Leaving API keys without spending limits, so a bug or leak runs up large charges.
- Picking a provider on headline price without checking rate limits and availability in your region.
Choose between GPU pods, serverless and clusters →Browse cloud providers for developers →Explore MLOps and LLMOps tools →
Decision resources
Continue the technical evaluation
Use the resources that match your decision stage. Profile data and linked sources were refreshed through November 10, 2026.
Comparisons
Ollama vs vLLM →Editorial guides
Runpod: choosing Pods, Serverless, or GPU Clusters →0 tools selectedCompare selected