Available models

A key can include 1–10 plan groups, and the models it can call are whatever that key's GET /v1/models returns — the union of its groups.

curl https://api.hop-base.com/v1/models \
  -H "Authorization: Bearer sk-your-key"

Chat and reasoning

Claude uses Anthropic Messages with the Base URL https://api.hop-base.com (no /v1); every other family uses the OpenAI compatible protocol with the Base URL https://api.hop-base.com/v1.

The tables on this page list the models on sale by chat, image, video, and speech; the Recommended column is the latest, everyday-first ID in each family. See the pricing catalog API for the full public catalog, and the model changelog for launches and retirements.

• Claude: claude-haiku-5-5, claude-sonnet-5-5, claude-opus-5-5, claude-fable-5, claude-opus-5, claude-opus-4-8, claude-opus-4-7, claude-opus-4-6, claude-sonnet-4-6, claude-haiku-4-5-20251001
• GPT: gpt-6-sol, gpt-6-luna, gpt-5.6-terra, gpt-5.5, codex-auto-review
• Gemini: gemini-4-argon, gemini-3.7-flash, gemini-3.6-flash, gemini-3.5-flash, gemini-3.5-flash-lite, gemini-3.1-pro-preview-customtools, gemini-3.1-flash-lite, gemini-3.1-flash-lite-preview, gemini-3-flash-preview, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite
• Qwen: qwen3.7-max, qwen3.7-plus, qwen3.7-flash
• DeepSeek: deepseek-v4-flash-202605
• Doubao Seed: doubao-seed-2-1-pro-260915, doubao-seed-2-1-lite-260915, doubao-seed-evolving, doubao-seed-character-260628, doubao-seed-2-0-mini-260428
• Grok: grok-4.6, grok-4.5, grok-4.3, grok-4.20-0309-reasoning, grok-4.20-multi-agent-0309

Doubao Seed notes: All seven doubao-seed-* models are in the Doubao Seed Cloud group and work over Chat Completions, Responses, and Anthropic Messages, with image input, tool calling, streaming (with usage), and context caching. Send images as a publicly downloadable URL or as base64. All except doubao-seed-character-260628 think by default and return the thinking in reasoning_content. doubao-seed-2-1-pro-260915 and doubao-seed-2-1-lite-260915 have a 1M context and up to 256K output; doubao-seed-2-1-pro-260628, doubao-seed-2-1-turbo-260628, doubao-seed-evolving, and doubao-seed-2-0-mini-260428 have a 256K context and up to 128K output; doubao-seed-character-260628 has a 128K context and up to 32K output. doubao-seed-evolving and doubao-seed-2-1-pro-260915 ignore tool_choice: with none or a named function they still decide on their own which tool to call, so pick doubao-seed-2-1-pro-260628 or doubao-seed-2-1-lite-260915 when you need to force or disable tool calls. doubao-seed-2-1-pro-260915 and doubao-seed-2-1-lite-260915 do not support logprobs; sending it returns an error, so leave it out.

Long-context billing

When a single request's input tokens (cached input included) go above the threshold, the whole request is billed at the multipliers below. Multipliers apply to the model's standard rates, which are shown in the signed-in model catalog. Caching does not shorten the input for tier purposes.

The two Gemini 3.1 Pro IDs carry these multipliers in the Gemini (all models, incl. image) group only. For per-family details, see OpenAI SDK, Grok, and Qwen.

doubao-seed-2-0-mini-260428 has two thresholds; the request is billed at the highest tier its input reaches (multipliers do not stack).

Images

Image generation and editing use the OpenAI Images protocol: POST /v1/images/generations and POST /v1/images/edits. Kling and Midjourney are the exceptions: they run as asynchronous tasks.

• GPT Image: gpt-image-2.5-sunburst, gpt-image-2
• Gemini Banana: gemini-nano-banana-2.1, gemini-3.1-flash-image-preview, gemini-3.1-flash-lite-image, gemini-2.5-flash-image
• Seedream: seedream-5-0-lite, seedream-4-5
• Grok Imagine: grok-imagine-image-quality, grok-imagine-image
• Kling: kling-image-v3-omni, kling-image-o1, kling-image-v2-1, kling-image-v2-1-i2i, kling-image-v2-1-multi-ref, kling-image-expand
• Midjourney: every task returns four images

Video

Endpoints and parameters differ by family; follow the guide for each one.

• Seedance: 2.0 standard / Fast / Mini tiers; take the complete IDs from GET /v1/models
• Kling: kling-v3, kling-v3-turbo, kling-o1, kling-v2-6, kling-v2-5-turbo, kling-v2-1, kling-v2-0, kling-v1-6, kling-v3-motion-control, kling-v2-6-motion-control, kling-avatar, kling-lip-sync
• MiniMax Hailuo: MiniMax-H3-Max, MiniMax-H3-SelfHosted
• Wan / HappyHorse: happyhorse-1.1-t2v, happyhorse-1.1-i2v, happyhorse-1.1-r2v

Speech

Text-to-speech is synchronous. MiniMax Speech is billed per character (a Han character counts 2); use the OpenAI-compatible POST /v1/audio/speech (input up to 4,096 characters) or the native POST /v1/t2a_v2 (up to 10,000). Doubao Speech uses only POST /v1/audio/speech (at most 5,000 characters per request for seed-tts-2.0 and 3,000 for seed-audio-1.0) and does not stream. For voice cloning with seed-icl-2.0, clone a voice with POST /v1/voice_clone first, then synthesize on the same endpoint.

Next steps

• Base URLs and protocols: which Base URL and protocol each client needs
• OpenAI SDK: Base URL, keys, and chat examples for the OpenAI protocol
• Troubleshooting: look up an error message