Meet the models.
Explore the possibilities.

claude-opus-5-5

Claude Opus 5.5 ↗

Anthropic’s Claude model for coding, research and everyday knowledge work. Explore what people are making with it, from interactive apps to motion design.

INPUT

$4 / 1M tokens

OUTPUT

$20 / 1M tokens

ROUTINE AGENT WORK

Claude Fable 5.5 ↗

An illustrated collection of creator demos attributed to Fable 5.5: animation, motion design and voxel worlds. The model identity and public release remain unconfirmed.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

deepseek/deepseek-v4.1-flash

DeepSeek V4.1 Flash 1017 ↗

A low-cost text model for research, extraction, classification, and other repeatable Agent tasks in Hermes.

INPUT

$0.265 / 1M tokens

OUTPUT

$1.059 / 1M tokens

Agent work

google/gemini-3.8-flash

Gemini 3.8 Flash ↗

A multimodal Agent model for research, coding, visual review, and structured work in Hermes, with clear task and cost guardrails.

INPUT

$0.675 / 1M tokens

OUTPUT

$3.375 / 1M tokens

Agent work

gemini-3.1-flash-image

Nano Banana 2 ↗

A fast image model for high-volume drafts and reference-image editing in a Hermes Agent workflow.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

Fast image drafts · reference-image editing · layout testing

gemini-3-pro-image

Nano Banana Pro ↗

A premium image model for product hero images, reference-image editing, and high-resolution visual work in a Hermes Agent workflow.

HERMES TEXT TO IMAGE

$0.10 / successful task

IMAGE EDIT

Live / see model page

Product hero images · reference edits · high-resolution visual delivery

gemini-nano-banana-2.1

Nano Banana 2.1 ↗

Gemini Nano Banana 2.1 is the Flash-level image model for fast drafts, multi-turn editing, and reference-heavy work in a Hermes Agent.

INPUT

$1.50 / 1M / 1M tokens

OUTPUT

$30 / 1M image tokens / 1M tokens

Fast generation · multi-turn edits · reference fusion · 1K–4K

—

Seedream 5.0 Pro ↗

Seedream 5.0 Pro, often searched as See Dream 5.0, is a design-focused image model for dense infographics, precise edits, layer separation, and multilingual commercial visuals.

INPUT

— / 1M tokens

OUTPUT

¥0.30 / image / 1M tokens

Infographics · precise edits · layer separation · multilingual design

suno-single-v5-5 / suno-custom-v5-5

Suno v5.5 ↗

A Hermes Agent guide to Suno v5.5: choose Single for fast song directions or Custom for controlled lyrics and style tags.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

Song ideation · custom lyrics · style control · asynchronous music jobs

suno-lyrics

Suno Lyrics ↗

A Hermes Agent guide to Suno Lyrics: plan structured, editable lyrics before sending them to a Suno Custom music job.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

Lyric planning · song structure · brand-safe drafts · Custom handoff

suno-single-v5 / suno-custom-v5

Suno v5 ↗

A Hermes Agent guide to Suno v5 for repeatable AI music and background-music production across content teams.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

Repeatable song jobs · batch BGM · saved prompts · v5.5 migration baseline

suno-single-v4-5 / suno-custom-v4-5

Suno v4.5 ↗

A Hermes Agent guide to Suno v4.5 for preserving legacy music workflows and testing a controlled migration.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

Legacy compatibility · rerun checks · long-form drafts · migration evidence

gemini-4-argon

Gemini 4 Argon ↗

A research guide for Google’s new flagship reasoning model: official API pricing, a shared 1M-token window, and the right role in a Hermes Agent workflow.

INPUT

$2.00 / 1M tokens

OUTPUT

$10.00 / 1M tokens

Complex reasoning · code review · multi-step Agents

qwen/qwen3.7-plus

Qwen3.7 Plus ↗

A Hermes Agent model for visual evidence, long source material, structured output, and controlled tool work.

INPUT

$0.295 / 1M tokens

OUTPUT

$1.177 / 1M tokens

Agent work

ROUTINE AGENT WORK

Kling 4.0 ↗

A controlled video model for Hermes Agent workflows that turn scripts, references, and keyframes into short production clips.

INPUT

— / 1M tokens

OUTPUT

— / 1M tokens

Controlled video workflows · references · keyframes

kimi-k3

Kimi K3 ↗

Kimi K3 is a strong choice for long coding, visual debugging and research that needs many steps. In Hermes Agent, use it when the task needs follow-through, not for every short request.

INPUT

$3.00 / 1M / 1M tokens

OUTPUT

$15.00 / 1M / 1M tokens

Use Kimi K3 when the work must keep going after the first answer.

glm-5

GLM-5 ↗

A general reasoning and coding model for Hermes Agent workflows.

INPUT

$1.00 / 1M / 1M tokens

OUTPUT

$3.20 / 1M / 1M tokens

General reasoning, coding, and tool use

glm-5-turbo

GLM-5-Turbo ↗

An OpenClaw focused model for tools, scheduled work, and persistent agent tasks.

INPUT

$1.03 / 1M / 1M tokens

OUTPUT

$3.82 / 1M / 1M tokens

OpenClaw, MCP, and persistent workflows

glm-5.1

GLM-5.1 ↗

A long horizon coding model for sustained autonomous engineering work.

INPUT

$1.40 / 1M / 1M tokens

OUTPUT

$4.40 / 1M / 1M tokens

Long horizon coding and autonomous delivery

glm-5.2

GLM-5.2 ↗

A project scale model for large repositories, 1M context, and cross file changes.

INPUT

$1.40 / 1M / 1M tokens

OUTPUT

$4.40 / 1M / 1M tokens

Large repositories and long context engineering

glm-5.3-flash

GLM-5.3-Flash ↗

A fast, lower cost multimodal model for visual coding, files, tools, and repeatable work.

INPUT

$0.15 / 1M / 1M tokens

OUTPUT

$0.50 / 1M / 1M tokens

Fast multimodal coding and batch automation

glm-5v-turbo

GLM-5V-Turbo ↗

A vision coding model for screenshots, GUI tasks, frontend recreation, and visual debugging.

INPUT

$1.03 / 1M / 1M tokens

OUTPUT

$3.82 / 1M / 1M tokens

Vision coding, GUI agents, and screenshot workflows