Pricing

Pricing is determined by attestation tier. Higher attestation reduces cost.

tier 1

Self-attested

Standard rate

Anyone with a HermesVault DID can register at the self-attested tier. Agent declares its own identity; no external verification required.

Suitable for: development, testing, internal tooling

tier 2

Runtime-signed

15% off standard

Recognized runtimes co-sign the agent identity at request time. The runtime attestation is verified against HermesBridge's registered framework public keys.

Recognized runtimes:

· LangChain

· AutoGen

· CrewAI

· Semantic Kernel

· custom OIDC-compatible

Suitable for: production agents, automated pipelines

tier 3

TEE-verified

30% off standard

Agent is measured inside a confidential compute enclave. Attestation quote is verified cryptographically by HermesBridge before any request is processed.

Supported hardware:

· Intel SGX (DCAP)

· AMD SEV-SNP

· AWS Nitro Enclaves

Suitable for: regulated workloads, sensitive data processing

Model pricing

Updated weekly — last: 2026-05-19

Prices shown are standard (self-attested) rates per million tokens. Runtime-signed agents receive 15% off; TEE-verified agents receive 30% off. Discounts apply to both input and output pricing.

model_id	input $/M	output $/M	context
claude-opus-4	$15.00	$75.00	200k
claude-sonnet-4	$3.00	$15.00	200k
gpt-4o	$2.50	$10.00	128k
gpt-4o-mini	$0.15	$0.60	128k
gemini-2.0-flash	$0.10	$0.40	1M
gemini-1.5-pro	$1.25	$5.00	2M
deepseek-v3	$0.27	$1.10	128k
qwen-2.5-72b	$0.40	$1.20	128k
mistral-large	$2.00	$6.00	128k
kimi-k1.5	$0.50	$2.50	128k
mimo-7b	$0.06	$0.24	32k

For volume commitments above $50K/month or custom attestation requirements, write to [email protected].