NEAR AI
Private, verifiable AI — Intel TDX + NVIDIA H200 TEE with on-chain attestation
NEAR AI Cloud runs inference inside Intel TDX confidential VMs backed by NVIDIA H200 Confidential Computing GPUs. TLS terminates inside the TEE (not at an external load balancer), so prompts are never exposed in plaintext outside the enclave. An optional E2EE layer lets clients encrypt messages with the model's attested public key before they leave the machine. Attestation reports are verifiable on-chain via the NEAR blockchain.
API Features
OpenAI Compatible ✓
Streaming ✓
Function Calling ✓
Batch Inference —
Vision —
Embeddings —
Models & Pricing
| Model | Family | Context | Max Out | Input /M | Output /M | Free |
|---|---|---|---|---|---|---|
| GPT-OSS 120B | OpenAI GPT | 128K | — | $0.15 | $0.55 | — |
| Whisper Large v3 | OpenAI GPT | 0K | — | $0.01 | $0.01 | — |
| Qwen3 30B A3B | Qwen | 256K | — | $0.15 | $0.55 | — |
| Qwen3.5 122B A10B | Qwen | 128K | — | $0.40 | $3.20 | — |
| Qwen3-Embedding 0.6B | Qwen | 40K | — | $0.01 | $0.01 | — |
| Qwen3-Reranker 0.6B | Qwen | 40K | — | $0.01 | $0.01 | — |
| Qwen3-VL 30B A3B | Qwen | 250K | — | $0.15 | $0.55 | — |
| GLM 5.1 | ZhipuAI GLM | 198K | — | $0.85 | $3.30 | — |
| GLM 5 | ZhipuAI GLM | 198K | — | $0.85 | $3.30 | — |