Guard provides the Claude models listed below. Choose the exact model and a compatible API; see billing for rates and discounts.
| id | Context | When | Category |
|---|---|---|---|
claude-opus-5 | 1M | flagship: complex code, agents | Premium |
claude-sonnet-5 | 1M | default: price/quality | General purpose |
claude-haiku-4-5 | 200K | fast and cheap | Lite |
claude-fable-5 | 1M | without an applied server prompt | Fable |
claude-fable-5-1 | 1M | without an applied server prompt | Fable |
Free-form spellings are mapped to canonical ids. Claude Code suffixes like claude-opus-4-8[1m] are stripped before the upstream.
// all of these → claude-opus-4-8
["opus 4.8", "Claude-Opus-4.8", "anthropic/claude-opus-4-8", "claude-opus-4-8[1m]"]For Kimi, the display name and API ID are separate. Copy the API ID into the model field without changes.
| Model | API ID | Context window |
|---|---|---|
| Kimi K3 | k3 | 1,048,576 tokens |
| Kimi K3-256K | k3-256k | 262,144 tokens |
| Kimi K2.7 Code | kimi-for-coding | 262,144 tokens |
| Kimi K2.7 Code HighSpeed | kimi-for-coding-highspeed | 262,144 tokens |
The context window covers the whole request together with the history and the model's answer. For k3 it is 1,048,576 tokens; for the other three models it is 262,144. What to take:
| Model | When to take it |
|---|---|
k3 | the largest window: a long history or a big document in one piece |
k3-256k | the same model with a 262,144 window and the same rates |
kimi-for-coding | code and agent loops, rates well below K3 |
kimi-for-coding-highspeed | the same thing, faster and exactly twice as expensive |
Kimi has no free-form spellings; the id is matched exactly. K3, kimi-k3 or Kimi K3 return 404 model_not_found before any charge, so copy the id from the table as it is.
curl https://api.guardrelay.ai/v1/models \
-H "Authorization: Bearer $GUARD_API_KEY"Two families, ten exact IDs in total: two Composer and eight Grok 4.6. Grok effort and the Fast variant are part of the ID; a separate field does not switch them.
| Model | API ID | Context | Images |
|---|---|---|---|
| Composer 2.5 | composer-2.5 | 200K | no |
| Composer 2.5 Fast | composer-2.5-fast | 200K | no |
| Grok 4.6 | cursor-grok-4.6-low, -medium, -high, -xhigh | 500K | yes |
| Grok 4.6 Fast | cursor-grok-4.6-low-fast, -medium-fast, -high-fast, -xhigh-fast | 500K | yes |
The effort level is chosen by the id itself, not by a request field: cursor-grok-4.6-low, -medium, -high, -xhigh. The level does not change the rate, so take low for a connection check.
The -fast suffix is a separate tariff: the same answer arrives faster and costs more. Grok has eight ids in total, four plain and four with -fast.
Composer has two ids: composer-2.5 and composer-2.5-fast. There are no effort levels, images are not accepted, and the context window is 200,000 tokens. The -fast variant is a separate tariff: the same answer arrives faster and costs more.
All ten ids are returned by the shared GET /v1/models. Copy the id into the model field without changes.