Developer docs
API playgroundTry for free, no card

Search company profiles

Featherless.ai

Full company profile

uuid0000p72

Namestring
Featherless.ai
Legal namestring
Featherless
Websiteurl
featherless.ai
Company typeenum
Private
Founded yearint
2023
Descriptiontext

Featherless.ai is a San Francisco-headquartered serverless AI inference provider, incorporated as a Delaware LLC and founded in 2023 (formerly known as Recursal). It offers an OpenAI-compatible REST API at api.featherless.ai that exposes 30,000+ open-source models across language, vision, audio, code, and multimodal modalities through a single endpoint, eliminating the need for customers to provision GPUs, configure inference servers, or manage quantization. Underlying the platform is a proprietary optimization stack spanning inference, model, and workflow optimization, with native integration on AMD's ROCm hardware via a strategic partnership, and research outputs from its wholly-owned Recursal Labs arm including the RWKV architecture (contributed to the Linux Foundation), the KVM novel attention architecture, and the RADLADS distillation method that converts large transformers to alternative-attention architectures at roughly 250x lower cost.

The company monetises primarily through flat-rate monthly subscriptions (Basic $10, Premium $25, Agent Standard $100, Agent Pro $200) with unlimited tokens and concurrency-based limits, billed via Stripe, alongside custom enterprise contracts for private deployments, SLAs, and multi-agent rollouts. Its go-to-market is a hybrid of self-serve API-first and product-led growth, supplemented by a community-led Builder Series that has sponsored 3,000+ open-source AI builders across 19 countries, and an emerging enterprise sales motion routed through HubSpot. Beyond raw inference, Featherless has expanded into a managed open-source agent marketplace (OpenClaw, Hermes Agent by Nous Research, NemoClaw) with sandboxing via Daytona, persistent storage, and flat-rate pricing targeted at 24/7 autonomous workflows. Customer segments include individual AI developers, open-source model creators (6,000+ hosted), agent developers, and enterprises with sovereignty or private-deployment requirements.

As of April 2026, Featherless has raised approximately $25 million across a $5M seed round (March 2025) and a $20M Series A co-led by AMD Ventures and Airbus Ventures with participation from BMW i Ventures, Kickstart Ventures, Panache Ventures, and Wavemaker Ventures. The company employs 51-100 staff distributed across San Francisco (HQ), Toronto, Singapore, and multiple European countries, with infrastructure spanning the EU and US. Co-founders are Eugene Cheah (CEO), Harrison Vanderbyl (CTO), and Wesley George (COO).

Short descriptiontext

Featherless.ai is a San Francisco-based serverless AI inference provider offering an OpenAI-compatible API that exposes 30,000+ open-source models across text, vision, audio, and code modalities to individual developers, open-source model creators, agent builders, and enterprise customers with sovereignty or private-deployment needs.

Operating statusenum
Operating
Ownership categoryenum
Headcount rangeband
11–50
akta.pro rankint
HeadquartersSan Francisco, United States
HQ citystring
San Francisco
HQ countrystring
United States
HQ regionstring
North America
Markets served

Serves global market

Offices4 records

Each record includes

City, Country, Type, Description, Source

Keyword5 values
serverless AI inference, LLM hosting platform, open-source model API, GPU orchestration, AI infrastructure
Industry3 codes
PRIMARYHDABAAAC
Serverless & Functions-as-a-Service (FaaS)
SECONDARYHDAEANAA
End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management)
Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise
SECONDARYHDAAACAF
Agents & Autonomous Workflows (Tool Use, Planning, Multi-Agent)
Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/Foundation Models & Generative AI (LLMs, multimodal)
NAICS code2 codes
  • Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services5182
  • Software Publishers513210
SIC code2 codes
  • Services-Prepackaged Software7372
  • Services-Computer Programming, Data Processing, Etc.7370
Product category
Cloud LLM Inference
GTM motion4 records

Each record includes

Type, Description, Source

Revenue model3 records
1Flat-rate monthly subscriptions
TypeSubscription Recurring
Description

Primary revenue stream: tiered monthly subscriptions (Basic $10, Premium $25, Agent Standard $100, Agent Pro $200) with unlimited tokens, billed via Stripe. Includes concurrency limits and context-window caps per tier. Renewals can be canceled any time; current month non-refundable.

featherless.ai
2Agent marketplace managed-runtime plans
TypeSubscription Recurring
Description

Subscription revenue from managed agent runtimes (OpenClaw, Hermes Agent, NemoClaw) bundled with inference, sandbox, persistent storage, and pricing — Standard $100/month, Pro $200/month, Enterprise custom.

featherless.ai
3Enterprise custom contracts
TypeManaged Services
Description

Custom enterprise engagements for private environments, regional sovereignty, production-grade assurance, dedicated resources, SLA, and multi-agent deployment.

featherless.ai
Marketing channels9 records

Each record includes

Title, Type, Stage, Description, Source

Distribution channels4 records

Each record includes

Title, Type, Scope, Target buyer, Description, Source

Cost components5 values
Technology or R&D, Infrastructure, Personnel, Operations, Marketing or Sales
Pricing details7 tiers
1Basic — $10/month, models up to 15B, 2 concurrent connections, up to 16K context
ModelSubscriptionBilling cadenceMonthly
Notes

$10.00 /month. Access to models up to 15B. Up to 2 concurrent connections. Up to 16K context. Unlimited tokens.

featherless.ai
2Premium — $25/month, access to DeepSeek/Kimi/GLM and any model size, 4 concurrent, up to 32K context
ModelSubscriptionBilling cadenceMonthly
Notes

$25.00 /month. Access to DeepSeek, Kimi and GLM. Access any model — no limit on size. Up to 4 concurrent connections. Up to 32K context. Unlimited tokens.

featherless.ai
3Agent Standard — $100/month, models up to 229B, 8 concurrent, 256K context, 1 agent runtime, standard sandbox, persistent storage
ModelSubscriptionBilling cadenceMonthly
Notes

$100.00 /month. Access any model up to 229B. Up to 8 concurrent connections. Up to 256K context. 1 agent runtime. Standard sandbox environment. Persistent Storage. Unlimited tokens.

featherless.ai
4Agent Pro — $200/month, any model no size limit, larger sandbox
ModelSubscriptionBilling cadenceMonthly
Notes

$200.00 /month. Access any model — no limit on size. Larger sandbox environment. Unlimited tokens.

featherless.ai
5OpenClaw Standard — $100/month, 1 agent runtime, full inference, standard sandbox, persistent storage
ModelSubscriptionBilling cadenceMonthly
Notes

$100/month. 1 agent runtime. Full inference access. Standard sandbox. Persistent storage.

featherless.ai
6OpenClaw Pro — $200/month, additional models, browser-use, larger sandbox
ModelSubscriptionBilling cadenceMonthly
Notes

$200/month. Everything in Standard. Additional Models. Browser-use capabilities. Larger Sandbox.

featherless.ai
7Enterprise — Custom pricing, multi-agent deployment, dedicated resources, enterprise security
ModelOtherBilling cadenceMulti-year contract
Notes

Custom. Multi-agent deployment. Dedicated resources. Custom Implementation. Enterprise-grade security.

featherless.ai
GTM typeB2B and B2C
B2B and B2C
Offering typeSoftware
Software
Brand1 of 3 records shown
1Recursal Labs
Description

The company's internal research group responsible for advancing foundational model architecture research, including the creation of the RWKV architecture and ongoing KVM (Key-Value Means) research.

featherless.ai
+2 more records
Core offering1 text field

Featherless.ai operates a serverless AI/LLM inference platform providing an OpenAI-compatible API at api.featherless.ai that gives developers single-endpoint access to 30,000+ open-source models without GPU provisioning or inference-server setup. The company sells tiered flat-rate monthly subscriptions ($10 Basic, $25 Premium, $100 Agent Standard, $200 Agent Pro, plus Enterprise custom) with unlimited tokens and concurrency-based rate limits, and additionally offers a managed runtime marketplace for open-source AI agents such as OpenClaw and Hermes Agent.

Differentiator
Functional benefit
Problem solved
Product and service3 records
1Featherless Serverless LLM Inference Platform
CategoryCloud LLM Inference
Description

Core open-source AI inference platform providing an OpenAI-compatible REST API at api.featherless.ai that exposes 30,000+ open-source LLM, vision, audio, and multimodal models through a single API key. Eliminates the need for users to provision GPUs, configure inference servers, or manage model quantization.

2Managed OpenClaw Runtime
CategoryManaged AI Agent Runtime
Description

Managed runtime for the open-source OpenClaw AI coding agent, providing a secure Daytona-powered sandboxed environment, persistent storage, 24/7 runtime, and access to the full 30,000+ model catalog with flat-rate monthly pricing.

3Hermes Agent on Featherless Marketplace
Scale indicator11 records

Each record includes

Type, Value, Description, Source

Partnership7 partners
Strategic tierFlagshipTypeStrategic or Co-development Partner
Description

Strategic partnership with AMD for hardware diversity and native integration on AMD's open-source ROCm platform. Through this partnership, significant open-source models run natively on AMD accelerators, enabling 'from model to metal' openness, lower inference costs, more available compute, and no lock-in. AMD Ventures also co-led the $20M Series A.

Strategic tierCoreTypeTechnology or Integration
Description

Featherless partnered with Daytona to build the sandbox layer for its managed agent runtime. Daytona's multi-layer container isolation provides each agent with a secure workspace and persistent storage so multi-day workflows can run unattended.

Strategic tierCoreTypeStrategic or Co-development Partner
Description

Nous Research's Hermes Agent is hosted in the Featherless agent marketplace as the first major third-party agent integration. Hermes Agent runs persistent, multi-session agents with 40+ built-in tools; on Featherless it has access to the full 30,000+ model catalog.

Strategic tierFlagshipTypeTechnology or Integration
Description

Featherless is positioned as the fastest-growing inference partner in the Hugging Face ecosystem. Hugging Face OAuth is offered as a sign-up method; supported models are sourced from the Hugging Face trending list and model creators publish to Featherless via Hugging Face collections (e.g., recursal/key-value-means).

Strategic tierCoreTypeTechnology or Integration
Description

OpenClaw is the first agent Featherless launched in its managed agent marketplace; it is an open-source AI coding agent (250,000+ GitHub stars, 50,000+ forks) hosted on Featherless inference.

Strategic tierFlagshipTypeStrategic or Co-development Partner
Description

Featherless's team created RWKV, a significant open-source foundational model architecture challenging transformer dominance, which is contributed to the Linux Foundation as a Linux Foundation project.

7Recursal (predecessor brand / research arm)
Strategic tierCoreTypeStrategic or Co-development Partner
Description

Featherless.ai was formerly known as Recursal. Recursal Labs continues as the company's research group publishing original research (RWKV, KVM, RADLADS) — feeds Featherless's optimization stack and differentiates the platform.

finsmes.com
Recent move7 records

Each record includes

Date, Type, Title, Description, Source

Expansion highlight6 records

Each record includes

Type, Description

Market position
Strengths4 records

Each record includes

Headline, Details, Source

Competitive moat6 records

Each record includes

Type, Details

Key risks7 records

Each record includes

Headline, Details, Source

Key highlights7 records

Each record includes

Headline, Details, Source

Customer concentration

Classification, Details

Named customers9 records

Each record includes

Name, Industry, Type, Use case, Source, UUID

Segment4 records

Each record includes

Title, Type, Primary, Description, Pain point addressed, Use case, Source

Ideal customer profile4 records

Each record includes

Profile, Firmographic size, Sales motion, Sales cycle length, Buying structure, Purchase trigger, Buyer persona, Geography, Industry vertical, Primary use case, Description, Pain points, Evidence proof points, Target buyer

Technology focused
Yes
API detail
Has APIbool
Yes

Docs URL, Description

Integration8 records

Each record includes

Title, Type, Description, Source

AI capability13 records

Each record includes

Type, Description, Source

AI maturity
App detail

Has app

Feature7 records

Each record includes

Title, Differentiator, Description, Source

Core technology
Revenue estimate
Valuation estimate
Number of profiles
Profiles3 records

Each record includes

Name, Designation, Designation category, Overview, Profile commentary, Source

Subsidiaries1 record

Each record includes

Name, Acquired on, Relationship type, Type, Business focus

Compliance3 records

Each record includes

Name, Class, Description

Funding overview

Funding stage, Last funding date, Total funding USD

Funding rounds4 records

Each record includes

Round, Amount USD, Date, Pre money valuation, Total investors, Investors, News

Investors15 records

Each record includes

Name, Type, Date of entry, Rounds participated, Website

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Each record includes

Name, Acquisition type, Announced date, Completed date, Status, Website, News

Investment

Each record includes

Name, Round, Announced date, Lead investor, Website, News

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Featherless.ai

Cloud LLM Inferencefeatherless.ai

Featherless.ai is a San Francisco-based serverless AI inference provider offering an OpenAI-compatible API that exposes 30,000+ open-source models across text, vision, audio, and code modalities to individual developers, open-source model creators, agent builders, and enterprise customers with sovereignty or private-deployment needs.

What Featherless.ai does

Featherless.ai is a San Francisco-headquartered serverless AI inference provider, incorporated as a Delaware LLC and founded in 2023 (formerly known as Recursal). It offers an OpenAI-compatible REST API at api.featherless.ai that exposes 30,000+ open-source models across language, vision, audio, code, and multimodal modalities through a single endpoint, eliminating the need for customers to provision GPUs, configure inference servers, or manage quantization. Underlying the platform is a proprietary optimization stack spanning inference, model, and workflow optimization, with native integration on AMD's ROCm hardware via a strategic partnership, and research outputs from its wholly-owned Recursal Labs arm including the RWKV architecture (contributed to the Linux Foundation), the KVM novel attention architecture, and the RADLADS distillation method that converts large transformers to alternative-attention architectures at roughly 250x lower cost.

The company monetises primarily through flat-rate monthly subscriptions (Basic $10, Premium $25, Agent Standard $100, Agent Pro $200) with unlimited tokens and concurrency-based limits, billed via Stripe, alongside custom enterprise contracts for private deployments, SLAs, and multi-agent rollouts. Its go-to-market is a hybrid of self-serve API-first and product-led growth, supplemented by a community-led Builder Series that has sponsored 3,000+ open-source AI builders across 19 countries, and an emerging enterprise sales motion routed through HubSpot. Beyond raw inference, Featherless has expanded into a managed open-source agent marketplace (OpenClaw, Hermes Agent by Nous Research, NemoClaw) with sandboxing via Daytona, persistent storage, and flat-rate pricing targeted at 24/7 autonomous workflows. Customer segments include individual AI developers, open-source model creators (6,000+ hosted), agent developers, and enterprises with sovereignty or private-deployment requirements.

As of April 2026, Featherless has raised approximately $25 million across a $5M seed round (March 2025) and a $20M Series A co-led by AMD Ventures and Airbus Ventures with participation from BMW i Ventures, Kickstart Ventures, Panache Ventures, and Wavemaker Ventures. The company employs 51-100 staff distributed across San Francisco (HQ), Toronto, Singapore, and multiple European countries, with infrastructure spanning the EU and US. Co-founders are Eugene Cheah (CEO), Harrison Vanderbyl (CTO), and Wesley George (COO).

Featherless.ai firmographics

Firmographics
Name
Featherless.ai
Legal name
Featherless
Website
https://featherless.ai
Company type
Private
Founded year
2023
Operating status
Operating
Headcount range
11–50 employees
Short description
Featherless.ai is a San Francisco-based serverless AI inference provider offering an OpenAI-compatible API that exposes 30,000+ open-source models across text, vision, audio, and code modalities to individual developers, open-source model creators, agent builders, and enterprise customers with sovereignty or private-deployment needs.
Ownership category
akta.pro rank

Featherless.ai industry classification

Industry
Product category
Cloud LLM Inference
NAICS
Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182), Software Publishers (513210)
SIC
Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370)
akta.pro industry
PRIMARYHDABAAAC
Serverless & Functions-as-a-Service (FaaS)
SECONDARYHDAEANAA
End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management)
Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise
SECONDARYHDAAACAF
Agents & Autonomous Workflows (Tool Use, Planning, Multi-Agent)
Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/Foundation Models & Generative AI (LLMs, multimodal)

Keywords

  • Serverless AI inference
  • LLM hosting platform
  • Open-source model API
  • GPU orchestration
  • AI infrastructure

Where Featherless.ai is headquartered

Location

Headquarters

HQ city
San Francisco
HQ country
United States
HQ region
North America

Offices4 records

Markets served

Featherless.ai business model

Business model
GTM type
B2B and B2C
Offering type
Software
Cost components
Technology or R&D, Infrastructure, Personnel, Operations, Marketing or Sales

Revenue model

  1. Flat-rate monthly subscriptions: Primary revenue stream: tiered monthly subscriptions (Basic $10, Premium $25, Agent Standard $100, Agent Pro $200) with unlimited tokens, billed via Stripe. Includes concurrency limits and context-window caps per tier. Renewals can be canceled any time; current month non-refundable.
  2. Agent marketplace managed-runtime plans: Subscription revenue from managed agent runtimes (OpenClaw, Hermes Agent, NemoClaw) bundled with inference, sandbox, persistent storage, and pricing — Standard $100/month, Pro $200/month, Enterprise custom.
  3. Enterprise custom contracts: Custom enterprise engagements for private environments, regional sovereignty, production-grade assurance, dedicated resources, SLA, and multi-agent deployment.

Pricing tiers

ModelBillingPrice
SubscriptionMonthlyBasic — $10/month, models up to 15B, 2 concurrent connections, up to 16K context
SubscriptionMonthlyPremium — $25/month, access to DeepSeek/Kimi/GLM and any model size, 4 concurrent, up to 32K context
SubscriptionMonthlyAgent Standard — $100/month, models up to 229B, 8 concurrent, 256K context, 1 agent runtime, standard sandbox, persistent storage
SubscriptionMonthlyAgent Pro — $200/month, any model no size limit, larger sandbox
SubscriptionMonthlyOpenClaw Standard — $100/month, 1 agent runtime, full inference, standard sandbox, persistent storage
SubscriptionMonthlyOpenClaw Pro — $200/month, additional models, browser-use, larger sandbox
OtherMulti-year contractEnterprise — Custom pricing, multi-agent deployment, dedicated resources, enterprise security

Go-to-market motion4 records

Distribution channels4 records

Marketing channels9 records

Featherless.ai product offering

Product offering

Core offering

Featherless.ai operates a serverless AI/LLM inference platform providing an OpenAI-compatible API at api.featherless.ai that gives developers single-endpoint access to 30,000+ open-source models without GPU provisioning or inference-server setup. The company sells tiered flat-rate monthly subscriptions ($10 Basic, $25 Premium, $100 Agent Standard, $200 Agent Pro, plus Enterprise custom) with unlimited tokens and concurrency-based rate limits, and additionally offers a managed runtime marketplace for open-source AI agents such as OpenClaw and Hermes Agent.

Differentiator

Problem solved

Functional benefit

Brands

  • Recursal Labs: The company's internal research group responsible for advancing foundational model architecture research, including the creation of the RWKV architecture and ongoing KVM (Key-Value Means) research.
  • OpenClaw
  • Phoenix

Products and services

  • Featherless Serverless LLM Inference Platform Core open-source AI inference platform providing an OpenAI-compatible REST API at api.featherless.ai that exposes 30,000+ open-source LLM, vision, audio, and multimodal models through a single API key. Eliminates the need for users to provision GPUs, configure inference servers, or manage model quantization.
  • Managed OpenClaw Runtime Managed runtime for the open-source OpenClaw AI coding agent, providing a secure Daytona-powered sandboxed environment, persistent storage, 24/7 runtime, and access to the full 30,000+ model catalog with flat-rate monthly pricing.
  • Hermes Agent on Featherless Marketplace

Companies that use Featherless.ai

Customer profile

Named customers9 records

Segments4 records

Ideal customer profiles4 records

Featherless.ai technology and API

Technology

Technology focussed Yes

API detail

Has API
Yes
API docs
API detail

Core technology

AI maturity

App detail

Integration8 records

AI capability13 records

Feature7 records

Featherless.ai partnerships and signals

Strategic signal

Partnerships

Seven partnerships are on record, tiered flagship and core.

  • AMD (AMD Ventures + ROCm)flagshipStrategic or Co-development PartnerStrategic partnership with AMD for hardware diversity and native integration on AMD's open-source ROCm platform. Through this partnership, significant open-source models run natively on AMD accelerators, enabling 'from model to metal' openness, lower inference costs, more available compute, and no lock-in. AMD Ventures also co-led the $20M Series A.
  • DaytonacoreTechnology or IntegrationFeatherless partnered with Daytona to build the sandbox layer for its managed agent runtime. Daytona's multi-layer container isolation provides each agent with a secure workspace and persistent storage so multi-day workflows can run unattended.
  • Nous ResearchcoreStrategic or Co-development PartnerNous Research's Hermes Agent is hosted in the Featherless agent marketplace as the first major third-party agent integration. Hermes Agent runs persistent, multi-session agents with 40+ built-in tools; on Featherless it has access to the full 30,000+ model catalog.
  • Hugging FaceflagshipTechnology or IntegrationFeatherless is positioned as the fastest-growing inference partner in the Hugging Face ecosystem. Hugging Face OAuth is offered as a sign-up method; supported models are sourced from the Hugging Face trending list and model creators publish to Featherless via Hugging Face collections (e.g., recursal/key-value-means).
  • OpenClawcoreTechnology or IntegrationOpenClaw is the first agent Featherless launched in its managed agent marketplace; it is an open-source AI coding agent (250,000+ GitHub stars, 50,000+ forks) hosted on Featherless inference.
  • Linux Foundation (RWKV)flagshipStrategic or Co-development PartnerFeatherless's team created RWKV, a significant open-source foundational model architecture challenging transformer dominance, which is contributed to the Linux Foundation as a Linux Foundation project.
  • Recursal (predecessor brand / research arm)coreStrategic or Co-development PartnerFeatherless.ai was formerly known as Recursal. Recursal Labs continues as the company's research group publishing original research (RWKV, KVM, RADLADS) — feeds Featherless's optimization stack and differentiates the platform.

Scale indicators11 records

Recent moves7 records

Expansion highlights6 records

Featherless.ai competitors and assessment

Company assessment

Market position

Strengths4 records

Competitive moat6 records

Key risks7 records

Key highlights7 records

Customer concentration

Featherless.ai social profiles

Digital presence

Featherless.ai compliance and trust

Trust signal

Compliance3 records

Featherless.ai financial estimates

Financial estimate

Revenue estimate

Valuation estimate

Featherless.ai leadership team

Management profile

Number of profiles

Profiles3 records

Featherless.ai subsidiaries and ownership

Company hierarchy

Subsidiaries1 record

Featherless.ai funding detail

Funding detail

Funding overview

Funding rounds4 records

Investors15 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

Featherless.ai M&A and investment

M&A and investment

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Featherless.ai

What does Featherless.ai do?

Featherless.ai operates a serverless AI/LLM inference platform providing an OpenAI-compatible API at api.featherless.ai that gives developers single-endpoint access to 30,000+ open-source models without GPU provisioning or inference-server setup. The company sells tiered flat-rate monthly subscriptions ($10 Basic, $25 Premium, $100 Agent Standard, $200 Agent Pro, plus Enterprise custom) with unlimited tokens and concurrency-based rate limits, and additionally offers a managed runtime marketplace for open-source AI agents such as OpenClaw and Hermes Agent.

Is Featherless.ai a public or private company?

Featherless.ai is a private company. It is classified as venture growth investor backed and is currently operating.

When was Featherless.ai founded?

Featherless.ai was founded in 2023. It employs 11 to 50 people.

Where is Featherless.ai based?

Featherless.ai is headquartered in San Francisco, United States, in the North America region.

How does Featherless.ai make money?

Three revenue lines are on record. Flat-rate monthly subscriptions are the primary driver. The others are agent marketplace managed-runtime plans and enterprise custom contracts.

Does Featherless.ai have an API?

Yes. Public OpenAI-compatible inference REST API hosted at https://api.featherless.ai/v1. Provides serverless access to 30,000+ open-source LLM models with a single API key. Authenticated via API key, supports OpenAI SDK and chat completions endpoint with model identifier substitution (e.g., model='zai-org/GLM-5.2'). No prompts or completions are logged. Plans impose concurrency-based rate limits (e.g., Basic: 2 concurrent, Premium: 4, Agent Standard: 8). Developer documentation is at featherless.ai/docs/overview.

What industry is Featherless.ai in?

Featherless.ai's product category is Cloud LLM Inference. Its primary akta.pro industry code is HDABAAAC, Serverless & Functions-as-a-Service (FaaS), with a secondary code of HDAEANAA, End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management). Its NAICS code is 5182 and its SIC code is 7372.

Unlock the full company data

50 free credits on sign-up, no credit card required.

Contact sales
Live signals
StartupHub.aiFeatherless AI Cuts $150K Token Bill to $10KFeatherless AI's Eugene Cheah said a customer cut its $150,000 monthly token bill to under $10,000 on a single node. The customer runs 2 trillion tokens daily at $0.30 per million tokens on 512 AMD MI325 GPUs. Cheah warned the optimization only applies if teams spend over $10,000 monthly on inference.The New StackWhy Featherless says you don’t need a tank to deliver a pizzaFeatherless released Simple Jev, an open-source library that converts open-source AI models into zero-shot classification engines with image support. It offers free public endpoints and production pricing starting at $0.03 per million input tokens, with output tokens free. The company expects hundreds of open-source clones.FeatherlessHow to use a Jev LLM classifier on Qwen3 with Simple JevSimple Jev is an open-source implementation of TypeSafe AI's Jev classifier, running on Qwen3.8-27B and Qwen3.6-35B-A3B on Featherless. It returns typed decisions with confidence scores, priced at $0.28–$0.30 per million input tokens with free output. The guide provides curl examples and production deployment details.FeatherlessDeepSeek-V4.1-Flash is live on FeatherlessDeepSeek launched V4.1-Flash on Featherless on September 10, a 763B-parameter model with 8B active at prefill and 16B at decode. It scores 90.6 on Terminal-Bench and 40 on Intelligence Index v4.3, and runs at 256K context with FP8 quantization.FeatherlessGLM-5.3-Flash is live on FeatherlessZ.ai launched GLM-5.3-Flash on Featherless, a 320B multimodal model under MIT license. It scores 84.3 on Terminal-Bench 2.1, beating Opus 4.8's 85.0, and runs at one-tenth the price. The model is available via OpenAI-compatible API with 256K context.FeatherlessHow to Use Kimi K3 in Opencode, Claude Code, and Your Coding AgentMoonshot released Kimi K3 weights, a 2.8T-parameter model with 1M-token context, and the guide shows how to integrate it into opencode, Claude Code, Cline, Roo Code, and Aider via Featherless's API. Benchmarks show K3 ranked #1 on Frontend Code Arena and 88.3 on Terminal-Bench 2.1, with serverless pricing at $2/M input and $10/M output.FeatherlessTransforming telecom with AI: Meet OTel 2.0AT&T, Microsoft, and AMD released OTel 2.0, an open large language model for telecom, live on Featherless. The model outperformed generalist models on the GSMA's TeleLogs benchmark and was downloaded over 25 million times before 2.0 shipped. Featherless sees this as the start of a proliferation of specialized AI models.The Next Web“How small can we go?” Interview with Eugene Cheah on AGI, open-source AI, and building a lean companyEugene Cheah, CEO of Featherless AI, argues AGI is an architectural problem and that reliability beats raw intelligence. The company raised a $20 million Series A co-led by AMD Ventures and Airbus Ventures, and hosts over 40,000 models. He aims to make AI accessible and believes open-source will eventually surpass closed models.FeatherlessHow to run Kimi K3 locally: hardware requirements and real costsMoonshot released Kimi K3 as open weights on 29 July 2026, a 2.8T parameter model with 104B active per token. Running it locally requires at least two H200 nodes for Q4 quantisation, costing about $740,000 in hardware or $35,000–$50,000 monthly in cloud rent. The guide recommends using Featherless's API instead.FeatherlessKimi K3 is live on FeatherlessMoonshot launched Kimi K3, an open-weight model with 2.8 trillion parameters and 104 billion active, on Featherless. It scores 88.3 on Terminal-Bench 2.1 and 93.5 on GPQA Diamond, competing with closed models. The model supports image input via MoonViT-V2 and runs on Featherless's API.