Empowering 10,000 builders and companies to build scalable, private, and safe intelligence.
San Francisco, CA
Joined August 2019
- Tweets6.4K
- Following835
- Followers141K
- Likes7.6K
Pinned Tweet
OPPO × Phala published a joint paper on Kubernetes Pod-level remote attestation for confidential AI.
Goal: prove the hardware, container image, and Pod identity before sensitive data enters the workload.
arxiv.org/abs/2606.03323
Phala Confidential AI: 105.50 billion tokens over the past 24 hours.
Top models:
GLM-5.3 Flash 43.47%
DeepSeek V4.1 Flash 23.98%
xing4.0-29b-a4b 7.70%
phala.com/models/z-ai/glm-5.…
TEE-backed AI infrastructure.
DeepSeek V4.1 Flash 1017 is live on Phala—served inside a TDX-attested GPU TEE.
Private, verifiable inference for long-horizon agents, visual understanding, reasoning.
phala.com/models/deepseek/de…
1M context. Tool use, structured outputs.
$0.75/M input · $1.6/M output.
docs.phala.com/phala-cloud/c…
Phala Confidential AI: 81.33 billion tokens over the past 24 hours.
Top models:
DeepSeek V4.1 Flash 56.30%
xing4.0-29b-a4b 12.22%
qwen3.8-27b 9.39%
phala.com/models/deepseek/de…
TEE-backed AI infrastructure.
Phala Confidential AI: 57.82 billion tokens over the past 24 hours.
Top models:
DeepSeek V4 Flash 0731 24.69%
GLM-5.3 Flash 24.69%
Kimi K3 14.28%
phala.com/models/deepseek/de…
TEE-backed AI infrastructure.
Private AI begins on-device.
Case study: how @solrouterai keeps its backend outside the plaintext path and opens encrypted prompts inside an attested Intel TDX enclave on Phala.
phala.com/posts/solrouter-pr…
Open-source @solrouterai SDK:
github.com/Router-Labs/route…
Docs:
docs.solrouter.com
A practical private AI blueprint: client-side encryption, attested workload keys, protected orchestration, and confidential inference.
Agent decisions often touch customer messages, policy documents, and approval state.
Kev-4B is now live in beta on Phala with NVIDIA Confidential Computing.
phala.com/models/jaredpalmer…
Kev-4B model-card benchmarks (self-reported, fp32):
• transfer-v4 locked OOD: 83.8% accuracy (656 questions)
• documents-v1 test: 90.3% (936)
• coverage at ≤5% error: 70.1% on transfer-v4 locked
Methodology: huggingface.co/jaredpalmer/k…
Kev-4B on Phala:
• Qwen3.5 4B Base
• TypeSafe System One compatible
• 8,192-token validated context
• $0.04/M input; free output
• NVIDIA Confidential Computing
• Beta
Read the launch article:
phala.com/posts/kev-4b-live-…
Agent commerce starts with verifiable trust.
Phala Cloud gives agents confidential runtimes and attestation linking their keys to code measurements.
@kudo_archi pairs that proof with Covenant NFTs recording agent commitments.
Onchain promises. Attested execution.
Kudo now uses @PhalaNetwork for agent attestation.
A Covenant NFT records what an agent promised. A Phala attestation proves what code is making that promise.
Promise, with proof.
Phala Confidential AI: 85.07 billion tokens over the past 24 hours.
Top models:
DeepSeek V4.1 Flash 51.54%
GLM-5.3 Flash 25.82%
Kimi K3 5.22%
phala.com/models/deepseek/de…
TEE-backed AI infrastructure.
Phala Confidential AI: 34.61 billion tokens over the past 24 hours.
Top models:
GLM-5.3 Flash 23.97%
DeepSeek V4.1 Flash 23.09%
GLM-5.3 14.66%
phala.com/models/z-ai/glm-5.…
TEE-backed AI infrastructure.
Congrats to @RIFTHQ on launching Rift 🎉 Phala and dstack give Rift a hardware-backed, verifiable execution layer for chaining complex onchain routes into a simple one-click experience. This is TEE infrastructure turning better execution into better UX.
Private AI needs a verifiable foundation.
Phala brings confidential inference and hardware-backed attestation to @AntSeed, helping protect sensitive code and documents during processing.
We build the infrastructure developers can build private AI on.
Antseed × Phala
Privacy shouldn’t mean taking a provider’s word for it. Your prompts can contain private code and documents.
@PhalaNetwork is live on Antseed, bringing hardware-backed TEE verification to our users.
Helping keep your data private.
Proud to power @orbiodotso’s Incognito with Phala’s confidential GPUs. End-to-end encrypted inference and hardware-attested execution, built right into agent workflows.
Introducing Orbio Incognito.
Agents can now make end-to-end encrypted inference requests through Orbio, powered by @PhalaNetwork’s confidential GPU infrastructure and hardware-attested execution.
Our agent toolkit handles encryption, key verification, verifiable receipts and local decryption across coding harnesses and SDKs.
Go beyond ZDR promises. Your prompts and responses stay encrypted through our gateway.
Your coding agent can read your entire codebase. Private AI Proxy verifies the AI service before that context leaves your machine, using hardware attestation, TLS key pinning, and fail-closed forwarding.
Download on Mac:
apps.apple.com/app/private-a…
Choose Phala, RedPill, or another compatible ACI service. The proxy checks fresh attestation evidence and pins TLS to the attested key before forwarding. Signed receipt audits add request-level evidence after delivery.
Read the ACI specification:
github.com/Dstack-TEE/privat…
Private AI Proxy is free and open source. Read how verification works before your coding agent sends a prompt:
phala.com/posts/private-ai-p…
Explore the code:
github.com/Dstack-TEE/privat…
