At a glance

Type
Direct provider
API base URL
https://api.featherless.ai/v1Source: API overview and common options
Billing
Plans. The Developer plan is billed per token from monthly credits (from $50 a month) and is the plan meant for API traffic. The Chat plan is a flat monthly fee for interactive use only.
Docs
https://featherless.ai/docs/api-overview-and-common-options
Prices
https://featherless.ai/pricing
Commercial relationship with Atlas
None.

Documented features

FeatureStatus per provider docsNotes
OpenAI-style chat completionsSupportedWorks with OpenAI SDKs by changing the base URL.
Public model listSupportedPublic GET /v1/models with prices and context per model.
Streaming (SSE)SupportedDocumented in the API docs.
Tool / function callingPer model or partialFlagged per model (tool_use) in the model list.
JSON mode / response_formatNot documented
Image inputPer model or partialFlagged per model (image_input) in the model list.
Not independently tested. Reliability, latency and uptime are not measured by Atlas.

What to know

  • Serving the exact Hugging Face repo id makes it possible to know which weights answer your request.
  • The model list reports a 32,768-token context for most refusal-reduced variants. The pricing page says the Developer plan allows "up to 256K", which may depend on the model.
  • The pricing page states "Prompts stored: Never" for every plan. Atlas has not verified this.

Refusal-reduced offerings

Provider and model idWhat is servedInput / 1MOutput / 1MCached input / 1MContextMax output100K in + 20K outChecked
Featherless
mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated
Names the exact Hugging Face repo$0.05$0.08not listed32,768Not stated$0.006614 November 2026
Featherless
coder3101/gemma-4-26B-A4B-it-heretic
Names the exact Hugging Face repo$0.40$0.80not listed32,76832,768$0.056014 November 2026
Featherless
llmfan46/gemma-4-26B-A4B-it-ultra-uncensored-heretic
Names the exact Hugging Face repo$0.40$0.80not listed32,76832,768$0.056014 November 2026
Featherless
p-e-w/Qwen3-4B-Instruct-2507-heretic
Names the exact Hugging Face repo$0.40$0.80not listed32,768Not stated$0.056014 November 2026
Featherless
mlabonne/gemma-3-27b-it-abliterated
Names the exact Hugging Face repo$0.40$1.20not listed32,768Not stated$0.064014 November 2026
Featherless
huihui-ai/Llama-3.3-70B-Instruct-abliterated
Names the exact Hugging Face repo$0.65$0.75not listed32,768Not stated$0.080014 November 2026
Featherless
dphn/Dolphin-Mistral-24B-Venice-Edition
Names the exact Hugging Face repo$0.70$1.16not listed32,768Not stated$0.093214 November 2026
Featherless
huihui-ai/Huihui-Qwen3.8-27B-abliterated
Names the exact Hugging Face repo$1.60$12.00not listed32,76832,768$0.400014 November 2026
USD per 1M tokens as published by each provider. Sorted by the cost of the 100K in + 20K out workload, lowest first; the workload column applies the same formula to every row and ignores caching.

Standard models in our data

Provider and model idWhat is servedInput / 1MOutput / 1MCached input / 1MContextMax output100K in + 20K outChecked
Featherless
google/gemma-4-26B-A4B-it
Standard model, not refusal-reduced$0.07$0.34not listed32,76832,768$0.013814 November 2026
Featherless
zai-org/GLM-4.7-Flash
Standard model, not refusal-reduced$0.06525$0.40not listed202,752Not stated$0.014514 November 2026
Featherless
meta-llama/Llama-3.3-70B-Instruct
Standard model, not refusal-reduced$0.65$0.75not listed32,768Not stated$0.080014 November 2026
Featherless
Qwen/Qwen3.8-27B
Standard model, not refusal-reduced$0.40$3.00not listed262,14432,768$0.100014 November 2026
USD per 1M tokens as published by each provider. Sorted by the cost of the 100K in + 20K out workload, lowest first; the workload column applies the same formula to every row and ignores caching.

Sources

  1. Featherless public model list (/v1/models), Featherless. Checked 14 November 2026.
  2. Pricing, Featherless. Checked 14 November 2026.
  3. API overview and common options, Featherless. Checked 14 November 2026.