Venice.ai

conditional free · inference_host · verification: partially_verified (60%, doc_fetch, 2026-07-17)

The record

Base URLs	https://api.venice.ai/api/v1 openai-compatible
API key required	yes
Credit card	no
Phone verification	unknown
May train on prompts	unknown
Quota	free_tier [rpm, tpm, credits] “Official rate-limit docs (default tier, text models by size class): XS "500 requests/min, 1,000,000 tokens/min"; S "75 requests/min, 750,000 tokens/min"; M "50 requests/min, 750,000 tokens/min"; L "20 requests/min, 500,000 tokens/min"; image 20 RPM; audio 60 RPM; embedding 500 RPM. "1 Diem = $1/day of compute". Free-LLM: "Daily limits for free tier \| Basic usage allowed \| Limits without Pro \| Resets daily".”
Signup friction	steps unknown
Env var	VENICE_API_KEY
Notes	Classified conditional_free: a free tier exists for the chat app, but sustained $0 API access effectively requires holding/staking the VVV token (Diem daily compute credits) — otherwise the API is pay-per-token via purchased credits. Model records below are therefore marked free: false. Catalog on the pricing page also lists proxied Claude/GPT/Gemini/DeepSeek variants alongside open-weights models.

Models (2)

model	capabilities	context	free	limits
llama-3.3-70b	chat	?	no	Priced $0.70 input / $2.80 output per 1M tokens on the official pricing page; rate limits apply per model-size class (text models L: "20 requests/min, 500,000 tokens/min" — size class of this model not confirmed). $0 usage only via VVV/Diem staking ("1 Diem = $1/day of compute").
venice-uncensored-1-2	chat	?	no	Priced "$0.20 input / $0.90 output" per 1M tokens on the official pricing page; rate limits apply per model-size class (size class of this model not confirmed). $0 usage only via VVV/Diem staking ("1 Diem = $1/day of compute").

Provenance

source_repo_freellm	curated_repo	trust: medium	checked 2026-07-17
source_venice_docs	official_docs	trust: high	checked 2026-07-17