RedPill

Build onConfidential AI.

Run confidential inference, bind the request to an attested gateway, verify the upstream provider, and keep a signed receipt for the response.

DocumentationGet API Key

Key Features for Developers

Smart Routing

Choose availability, profit, or e2ee routing. E2EE mode never downgrades a confidential backend to a non-TEE fallback.

Virtual Keys

Create sk-rp-* keys with per-key budgets, RPM and TPM limits, and routing strategy controls for each app or customer.

Attestation & Receipts

Call /v1/attestation/report with a nonce, then fetch the signed receipt through /v1/receipt/{id}.

End-to-End Encryption

E2EE v2 uses AES-256-GCM, secp256k1 key exchange, nonce binding, and timestamp replay protection.

Attested Audit Sessions

Receipts reference read-only audit sessions so relying parties can check target, verifier, evidence digest, and channel binding.

Full API Surface

Chat, completions, embeddings, and model discovery work the same way you already call them; RedPill middleware extends the broader API surface.

Browse Models
Up and running in minutes

YOUR SDK.
VERIFIABLE AI.

Keep the OpenAI client you already use. Change the base URL, choose a confidential model, then verify the gateway attestation, the response receipt, and the attested session.

View Full Docs
chat.py
import osfrom openai import OpenAIclient = OpenAI(    base_url="https://api.redpill.ai/v1",    api_key=os.environ["REDPILL_API_KEY"],)response = client.chat.completions.create(    model="openai/gpt-oss-120b",    messages=[        {"role": "user", "content": "Hello, confidential world"}    ],)print(response.id)print(response.choices[0].message.content)

Open Source Where It Matters

Audit the gateway, run the verifier, inspect the chat UX, and track the ACI signed-receipts spec as a developer-preview proof layer.

Open Source Verification

Explore AI Models

From private models in GPU TEE to all your favorites.

z-ai logo
Z.ai: GLM 5.2
GPU TEE
GLM-5.2 is Z.ai's flagship model for the era of long-horizon tasks. With a truly usable 1M-token context window, it can handle project-level engineering context and execute long-running tasks more reliably. Served as a text-only TEE deployment via Phala.
by Phala|1M context|$1.40/M input|$4.40/M output|$0.70/M cache read
Intel TDXNVIDIA CC
google logo
Google: Gemma 4 31B
GPU TEE
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense model. Features a 256K token context window, configurable thinking/reasoning mode, native function calling, and strong multilingual performance. Served as a text-only TEE deployment via NEAR AI.
by Phala|262K context|$0.15/M input|$0.46/M output|$0.075/M cache read
Intel TDXNVIDIA CC
phala logo
Phala: Gemma-4 26B-A4B Uncensored (Heretic)
GPU TEE
Uncensored "Heretic" variant of google/gemma-4-26B-A4B-it created using Heretic v1.2.0 with the Arbitrary-Rank Ablation (ARA) method and row-norm preservation. Refusals drop from 100/100 to 11/100 with KL divergence 0.0499 vs the base model. The base Gemma 4 26B A4B is a Mixture-of-Experts model with 25.2B total / 3.8B active parameters (8 active / 128 total experts), 30-layer transformer with hybrid local sliding (1024) + global attention, supporting a 256K context window. Natively multimodal (text + images, variable aspect ratios). Strong on coding, reasoning, function calling, with native system prompt support across 35+ languages. Served on Phala in TDX-attested H200 enclave with end-to-end ECDSA response signing; vLLM-compatible FP8-Static quantization by cloud19 (router excluded from quantization).
by Phala|66K context|$0.15/M input|$0.70/M output
Intel TDXNVIDIA CC
phala logo
Phala: Qwen3.6 35B-A3B Uncensored (Aggressive)
GPU TEE
Uncensored "Aggressive" variant of Qwen3.6-35B-A3B from Alibaba's Qwen team. The fine-tune by HauhauCS removes refusal behaviors (0/465 refusals) without modifying datasets or core capabilities. The base architecture is a 35B-parameter Mixture-of-Experts model with 256 experts routing 8 per token (~3B active params), 40 layers, and a hybrid linear+full-softmax attention mechanism (3:1 ratio). Supports a native 262K context and is natively multimodal across text, images, and video. Served on Phala in TDX-attested H200 enclave with end-to-end ECDSA response signing; FP8 quantization by lamianlbe.
by Phala|131K context|$0.30/M input|$1.50/M output
Intel TDXNVIDIA CC
phala logo
Phala: Venice Uncensored 24B
GPU TEE
Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai. This model is designed as an “uncensored” instruct-tuned LLM, preserving user control over alignment, system prompts, and behavior. Intended for advanced and unrestricted use cases, Venice Uncensored emphasizes steerability and transparent behavior, removing default safety and alignment layers typically found in mainstream assistant models.
by Phala|33K context|$0.20/M input|$0.90/M output
Intel TDXNVIDIA CC
Will be deprecated on Jul 29, 2026 18:00
sentence-transformers logo
Sentence Transformers: all-MiniLM-L6-v2
GPU TEE
The all-MiniLM-L6-v2 embedding model maps sentences and short paragraphs into a 384-dimensional dense vector space, enabling high-quality semantic representations that are ideal for downstream tasks such as information retrieval, clustering, similarity scoring, and text ranking.
by Phala|512 context|$0.005/M input|$0.00/M output
Intel TDXNVIDIA CC
View All Models

Start Building.

API Documentation

Comprehensive guides, API references, and tutorials to help you integrate RedPill into your applications. Try the interactive playground or get a free API key.

View DocumentationGet API Key

Developer Community

Join our Discord community to connect with other developers, get help with integration questions, and share what you're building with RedPill.

Join DiscordGitHub

Get an API key in 2 minutes

Start with the OpenAI SDK, choose a confidential model, and add verifier checks when you need audit-grade proof.

Get API Key
Private Chat
TEE model
AI
Hi. Ask me anything and I'll open the confidential chat with openai/gpt-oss-120b selected.
Zero data retentionVerifier ready
RedPill

The confidential AI cloud: verifiable inference with attestation reports, signed receipts, audit sessions, and E2EE paths.

© 2026 Hashforest Technology LLC. All rights reserved.