mirror of https://github.com/whit3rabbit/anyllm-proxy.git synced 2026-09-23 00:00:51 +00:00

Files

T

whit3rabbitandClaude Sonnet 4.6 545bdbbbd3 feat: add bedrock native passthrough, generic passthrough, provider docs, and managed backend fixes

Adds bedrock_native.rs (Converse/InvokeModel with SigV4) and
generic_passthrough.rs catch-all for Translate mode. Adds comprehensive
provider reference docs (docs/providers/, docs/ENDPOINTS.md). Fixes
managed backend admin UI (BackendForm, ManagedBackendsSection) and
admin route/model handler issues. Adds automated model pricing update
workflow (scripts/update_pricing.py, .github/workflows/update-pricing.yml).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-08 16:36:16 -05:00

3.5 KiB

Raw Permalink Blame History

HuggingFace

HuggingFace Inference, covering both serverless inference (Inference API) and dedicated Inference Endpoints (TGI/vLLM-backed deployments).

LiteLLM prefix: huggingface/
Status: Stub — routes through OpenAI-compatible client
Docs: https://huggingface.co/docs/api-inference/en/index

Authentication

Variable	Required	Description
`HUGGINGFACE_API_KEY`	Yes (one of)	HuggingFace user access token
`HF_TOKEN`	Yes (one of)	Alias for `HUGGINGFACE_API_KEY`; either is accepted
`OPENAI_BASE_URL`	Situational	Required for dedicated Inference Endpoints (see Notes)

Quick Start

Single-Backend (env vars)

Serverless inference (public models on the Inference API):

BACKEND=huggingface \
  HF_TOKEN=hf_your-token \
  OPENAI_BASE_URL=https://api-inference.huggingface.co/models/meta-llama/Meta-Llama-3.1-8B-Instruct/v1 \
  cargo run -p anyllm_proxy
# Docker:
docker run \
  -e BACKEND=huggingface \
  -e HF_TOKEN=hf_your-token \
  -e OPENAI_BASE_URL=https://api-inference.huggingface.co/models/meta-llama/Meta-Llama-3.1-8B-Instruct/v1 \
  -e PROXY_OPEN_RELAY=true \
  -p 3000:3000 followthewhit3rabbit/anyllm-proxy

LiteLLM YAML Config

Dedicated Inference Endpoint (per-deployment URL):

model_list:
  - model_name: llama-3.1-8b
    litellm_params:
      model: huggingface/meta-llama/Meta-Llama-3.1-8B-Instruct
      api_key: "env:HF_TOKEN"
      api_base: "https://<endpoint-id>.endpoints.huggingface.cloud/v1"

Serverless Inference API:

model_list:
  - model_name: llama-3.1-8b-serverless
    litellm_params:
      model: huggingface/meta-llama/Meta-Llama-3.1-8B-Instruct
      api_key: "env:HF_TOKEN"
      api_base: "https://api-inference.huggingface.co/models/meta-llama/Meta-Llama-3.1-8B-Instruct/v1"

Usage Examples

Anthropic Messages API

curl http://localhost:3000/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: $PROXY_API_KEYS" \
  -d '{"model": "meta-llama/Meta-Llama-3.1-8B-Instruct", "max_tokens": 1024, "messages": [{"role": "user", "content": "Hello"}]}'

OpenAI Chat Completions API

curl http://localhost:3000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $PROXY_API_KEYS" \
  -d '{"model": "meta-llama/Meta-Llama-3.1-8B-Instruct", "messages": [{"role": "user", "content": "Hello"}]}'

Capabilities

Feature	Supported
Chat Completions	✓
Streaming	✓
Tool Use	—
Embeddings	✓
Vision	—
Batch	—

Notable Models

Model ID	Context	Notes
`meta-llama/Meta-Llama-3.1-8B-Instruct`	128k	Llama 3.1 8B, commonly available serverless
`meta-llama/Meta-Llama-3.1-70B-Instruct`	128k	Llama 3.1 70B, requires PRO or dedicated endpoint
`mistralai/Mistral-7B-Instruct-v0.3`	32k	Mistral 7B

Notes

HuggingFace has no single shared base URL. There are two deployment types:

Serverless Inference API: https://api-inference.huggingface.co/models/<org>/<model>/v1. Available for popular gated and public models; rate-limited on free tier; requires accepting model terms on huggingface.co.
Dedicated Inference Endpoints: https://<endpoint-id>.endpoints.huggingface.cloud/v1. Per-deployment URL created in the HuggingFace dashboard. Pay-per-hour pricing with guaranteed capacity.

Always set OPENAI_BASE_URL or per-model api_base in the YAML config; the default base URL is empty. Tool use support depends on the specific model and TGI version deployed.

3.5 KiB Raw Permalink Blame History