LLM7 model guide

Hermes Agent with LLM7

Configure Hermes Agent with LLM7: eligible models, setup instructions, and a copyable configuration.

Requirements and limitations

Requires reported tool calling and at least 64,000 context tokens. Auxiliary models and individual tools may need separate configuration.

Setup

Install Hermes using its documentation, then run hermes model and select Custom endpoint.
Use the configuration below in ~/.hermes/config.yaml and replace the API key placeholder.
Start a new session after changing providers. Budget tools and conversation within the published context window.

Configuration model

model:
  provider: custom
  default: "DeepSeek-V4-Flash-0731"
  base_url: https://api.llm7.io/v1
  api_key: "YOUR_LLM7_API_KEY"
  context_length: 400000

Sort models

Eligible models and published pricing
Model	Pricing	Context	Tool calling
DeepSeek-V4-Flash-0731	$0.02 USD input and $0.04 USD output per 1M tokens	400,000	Supported
Inkling-Small	$0.50 USD input and $1.20 USD output per 1M tokens	512,000	Supported
XiaomiMiMo/MiMo-V2.5	$0.40 USD input and $2.00 USD output per 1M tokens	256,000	Supported
XiaomiMiMo/MiMo-V2.5-Pro	$1.00 USD input and $3.00 USD output per 1M tokens	1,024,000	Supported
claude-fable-5	$4.50 USD input and $30.00 USD output per 1M tokens	1,000,000	Supported
claude-fable-5-1	$4.00 USD input and $15.00 USD output per 1M tokens	1,000,000	Supported
claude-haiku-4-5	$0.04 USD input and $0.20 USD output per 1M tokens	200,000	Supported
claude-opus-4-8	$2.25 USD input and $12.19 USD output per 1M tokens	1,000,000	Supported
claude-opus-5	$2.50 USD input and $12.50 USD output per 1M tokens	1,000,000	Supported
claude-sonnet-4-6	$0.12 USD input and $0.45 USD output per 1M tokens	1,000,000	Supported
claude-sonnet-5	$0.45 USD input and $2.25 USD output per 1M tokens Minimum: $0.000784/request	1,000,000	Supported
deepseek-v4-flash:0731	$0.06 USD input and $0.18 USD output per 1M tokens	1,024,000	Supported
deepseek-v4-flash_0731	$0.18 USD input and $0.60 USD output per 1M tokens	1,000,000	Supported
deepseek-v4-pro	$1.11 USD input and $3.33 USD output per 1M tokens	1,000,000	Supported
gemini-3-flash	$0.03 USD input and $0.08 USD output per 1M tokens	1,048,576	Supported
gemini-3.1-flash-lite	$0.02 USD input and $0.04 USD output per 1M tokens	256,000	Supported
gemini-3.7-flash	$0.06 USD input and $0.30 USD output per 1M tokens	1,000,000	Supported
gemini-3.8-flash-high	$0.05 USD input and $0.15 USD output per 1M tokens	1,000,000	Supported
gemma4:31b	$0.07 USD input and $0.23 USD output per 1M tokens	262,000	Supported
glm-5.3	$0.50 USD input and $2.00 USD output per 1M tokens	1,000,000	Supported
glm-5.3-flash	$0.15 USD input and $0.55 USD output per 1M tokens	1,048,576	Supported
gpt-5.5	$0.20 USD input and $1.00 USD output per 1M tokens Minimum: $0.0001/request	1,050,000	Supported
gpt-5.6-sol	$1.92 USD input and $5.40 USD output per 1M tokens	1,000,000	Supported
gpt-5.6-terra	$0.24 USD input and $1.00 USD output per 1M tokens	1,000,000	Supported
gpt-6-astra	$5.00 USD input and $10.00 USD output per 1M tokens	1,000,000	Supported
grok-4.5	$0.30 USD input and $1.00 USD output per 1M tokens	500,000	Supported
grok-4.6	$0.40 USD input and $0.50 USD output per 1M tokens	500,000	Supported
kimi-k3	$2.00 USD input and $10.00 USD output per 1M tokens	1,000,000	Supported
llama-4-maverick	$0.19 USD input and $0.75 USD output per 1M tokens	1,048,576	Supported
minimax-m2.7	$0.03 USD input and $0.05 USD output per 1M tokens	180,000	Supported
seed-2.0-mini	$0.10 USD input and $0.40 USD output per 1M tokens	250,000	Supported