Skip to content
SayGMDocs
Dashboard

Retrieve a model

GET
/models/{model}
curl --request GET \
--url https://api.saygm.com/v1/models/gpt-5.5

Returns one canonical buyer model entry. Authentication is optional; with Authorization, the gateway validates the key to include visible composites. Catalog responses include ETag and Cache-Control headers; send If-None-Match to revalidate.

model
required
string

A model id exactly as returned by /v1/models. Percent-encode / in ids that contain a slash.

Examples

Frontier model

gpt-5.5
If-None-Match
string

Revalidate the representation using its ETag.

The catalog entry for the requested model, with ETag and Cache-Control headers.

Media typeapplication/json

One catalog entry. Direct and cascade products have a pricing card; fusion products omit pricing and price_range and carry billing and fusion metadata.

object
id
required

Canonical buyer model id to send in a request.

string
object
required
string
Allowed value: model
created
required

Fixed Unix timestamp retained for OpenAI SDK compatibility.

integer
owned_by
required

Organization credited with the model.

string
tier
required

Trust boundary, derived from confidential and model_class: confidential is in-TEE inference; open and frontier are served by an upstream provider under that provider’s own terms. Unreviewed models report frontier.

string
Allowed values: frontier confidential open
model_class

Reviewed weights class. Omitted until GM has reviewed the model; an omitted field claims nothing.

string
Allowed values: open frontier
confidential

true when inference runs inside a TEE. Omitted until GM has reviewed the model; an omitted field claims nothing.

boolean
display_name

Curated human-readable name. Omitted when GM has none, and for composites.

string
gateway_provider
required

GM’s provider namespace for the model, usually its maker. It does not identify the upstream service that receives a request; a model can be served by several upstream providers, which X-GM-Providers can restrict. Composite products report gm.

string
api_shapes
required
Array<string>
Allowed values: chat.completions responses messages generateContent streamGenerateContent images.generations videos
reasoning_efforts

Reasoning effort values supported by this model.

Array<string>
reasoning_effort_aliases

Accepted reasoning effort aliases and their native values.

object
key
additional properties
string
pricing

One rate card for direct and cascade products. Omitted for fusion products because a fusion is billed per leg.

object
unit
required

Pricing unit: per million tokens for token models, or per generated video second for video models.

string
Allowed values: ndollars_per_mtok ndollars_per_video_second
currency
required
string
Allowed value: USD
basis
required
string
Allowed values: cheapest_eligible_offer published_retail
dimensions
required

Rates in nano-dollars. Per-token dimensions are per million tokens; output_per_image_ndollars is per generated image; output_per_video_second_ndollars is per generated video second. Dimensions are omitted when not offered and never serialized as null.

object
input_per_mtok_ndollars
integer
output_per_mtok_ndollars
integer
cache_read_per_mtok_ndollars
integer
cache_write_per_mtok_ndollars
integer
cache_write_5m_per_mtok_ndollars
integer
cache_write_1h_per_mtok_ndollars
integer
audio_input_per_mtok_ndollars
integer
audio_output_per_mtok_ndollars
integer
image_input_per_mtok_ndollars
integer
image_output_per_mtok_ndollars
integer
output_per_image_ndollars
integer
cache_storage_per_mtok_hour_ndollars
integer
long_context_threshold_tokens
integer
long_context_input_per_mtok_ndollars
integer
long_context_output_per_mtok_ndollars
integer
long_context_cache_read_per_mtok_ndollars
integer
long_context_cache_write_per_mtok_ndollars
integer
output_per_video_second_ndollars

Video output price per generated second in nano-dollars.

integer
long_context_cache_write_5m_per_mtok_ndollars
integer
long_context_cache_write_1h_per_mtok_ndollars
integer
surcharges
required
object
key
additional properties
One of:
object
kind
required
Allowed value: per_event
unit_ndollars
required
integer
price_range

Direct-product bounds. Omitted for composites without a single rate card; optional fields are omitted, never null.

object
unit
required

Pricing unit: per million tokens for token models, or per generated video second for video models.

string
Allowed values: ndollars_per_mtok ndollars_per_video_second
currency
required
string
Allowed value: USD
floor
object
basis
required
string
Allowed values: cheapest_eligible_offer highest_eligible_offer published_retail
dimensions
required

Rates in nano-dollars. Per-token dimensions are per million tokens; output_per_image_ndollars is per generated image; output_per_video_second_ndollars is per generated video second. Dimensions are omitted when not offered and never serialized as null.

object
input_per_mtok_ndollars
integer
output_per_mtok_ndollars
integer
cache_read_per_mtok_ndollars
integer
cache_write_per_mtok_ndollars
integer
cache_write_5m_per_mtok_ndollars
integer
cache_write_1h_per_mtok_ndollars
integer
audio_input_per_mtok_ndollars
integer
audio_output_per_mtok_ndollars
integer
image_input_per_mtok_ndollars
integer
image_output_per_mtok_ndollars
integer
output_per_image_ndollars
integer
cache_storage_per_mtok_hour_ndollars
integer
long_context_threshold_tokens
integer
long_context_input_per_mtok_ndollars
integer
long_context_output_per_mtok_ndollars
integer
long_context_cache_read_per_mtok_ndollars
integer
long_context_cache_write_per_mtok_ndollars
integer
output_per_video_second_ndollars

Video output price per generated second in nano-dollars.

integer
long_context_cache_write_5m_per_mtok_ndollars
integer
long_context_cache_write_1h_per_mtok_ndollars
integer
route_ceiling
object
basis
required
string
Allowed values: cheapest_eligible_offer highest_eligible_offer published_retail
dimensions
required

Rates in nano-dollars. Per-token dimensions are per million tokens; output_per_image_ndollars is per generated image; output_per_video_second_ndollars is per generated video second. Dimensions are omitted when not offered and never serialized as null.

object
input_per_mtok_ndollars
integer
output_per_mtok_ndollars
integer
cache_read_per_mtok_ndollars
integer
cache_write_per_mtok_ndollars
integer
cache_write_5m_per_mtok_ndollars
integer
cache_write_1h_per_mtok_ndollars
integer
audio_input_per_mtok_ndollars
integer
audio_output_per_mtok_ndollars
integer
image_input_per_mtok_ndollars
integer
image_output_per_mtok_ndollars
integer
output_per_image_ndollars
integer
cache_storage_per_mtok_hour_ndollars
integer
long_context_threshold_tokens
integer
long_context_input_per_mtok_ndollars
integer
long_context_output_per_mtok_ndollars
integer
long_context_cache_read_per_mtok_ndollars
integer
long_context_cache_write_per_mtok_ndollars
integer
output_per_video_second_ndollars

Video output price per generated second in nano-dollars.

integer
long_context_cache_write_5m_per_mtok_ndollars
integer
long_context_cache_write_1h_per_mtok_ndollars
integer
ceiling
required
object
basis
required
string
Allowed values: cheapest_eligible_offer highest_eligible_offer published_retail
dimensions
required

Rates in nano-dollars. Per-token dimensions are per million tokens; output_per_image_ndollars is per generated image; output_per_video_second_ndollars is per generated video second. Dimensions are omitted when not offered and never serialized as null.

object
input_per_mtok_ndollars
integer
output_per_mtok_ndollars
integer
cache_read_per_mtok_ndollars
integer
cache_write_per_mtok_ndollars
integer
cache_write_5m_per_mtok_ndollars
integer
cache_write_1h_per_mtok_ndollars
integer
audio_input_per_mtok_ndollars
integer
audio_output_per_mtok_ndollars
integer
image_input_per_mtok_ndollars
integer
image_output_per_mtok_ndollars
integer
output_per_image_ndollars
integer
cache_storage_per_mtok_hour_ndollars
integer
long_context_threshold_tokens
integer
long_context_input_per_mtok_ndollars
integer
long_context_output_per_mtok_ndollars
integer
long_context_cache_read_per_mtok_ndollars
integer
long_context_cache_write_per_mtok_ndollars
integer
output_per_video_second_ndollars

Video output price per generated second in nano-dollars.

integer
long_context_cache_write_5m_per_mtok_ndollars
integer
long_context_cache_write_1h_per_mtok_ndollars
integer
billing

Present only for fusion products; omitted for direct and cascade products.

object
mode
required
string
Allowed value: metered_per_leg
fusion

Present only for fusion products.

object
members
required
Array<object>
object
provider
required
string
model
required
string
available
required
boolean
adjudicator
required
object
provider
required
string
model
required
string
available
required
boolean
strategy
required
string
Allowed values: rank_then_fuse synthesize
member_count
required
integer
available
required

Whether the gateway is currently accepting inference requests for this model.

boolean
coming_soon
required

true for an announced model that cannot be called yet. Such a model is unavailable and has no pricing or price_range.

boolean
capabilities
required

Hard capability flags and any published image or video generation limits.

object
vision
required
boolean
tools
required
boolean
image_input
required
boolean
image_output
required
boolean
image_generation
required
boolean
api

Image API that serves the product, for example openai_images or gemini_generate_content.

string
Allowed values: gemini_generate_content openai_images openai_videos
max_images_per_request
integer
max_reference_images
integer
supported_aspect_ratios
Array<string>
video_generation
required
boolean
supported_frame_roles
Array<string>
Allowed values: first_frame last_frame
last_frame_requires_first_frame
boolean
video_seconds_default
integer
>= 4 <= 15
supported_video_seconds
Array<integer>
>= 1 items <= 12 items
supported_video_sizes
Array<string>
>= 1 items <= 12 items
artifact_ttl_seconds
integer
>= 60
Example
{
"id": "gpt-5.5",
"object": "model",
"owned_by": "anthropic",
"tier": "frontier",
"model_class": "open",
"gateway_provider": "anthropic",
"api_shapes": [
"chat.completions"
],
"pricing": {
"unit": "ndollars_per_mtok",
"currency": "USD",
"basis": "cheapest_eligible_offer",
"surcharges": {
"additionalProperty": {
"kind": "per_event"
}
}
},
"price_range": {
"unit": "ndollars_per_mtok",
"currency": "USD",
"floor": {
"basis": "cheapest_eligible_offer"
},
"route_ceiling": {
"basis": "cheapest_eligible_offer"
},
"ceiling": {
"basis": "cheapest_eligible_offer"
}
},
"billing": {
"mode": "metered_per_leg"
},
"fusion": {
"strategy": "rank_then_fuse"
},
"capabilities": {
"api": "gemini_generate_content",
"supported_frame_roles": [
"first_frame"
]
}
}
Cache-Control
string
Allowed value: no-cache, stale-if-error=300

Model catalog revalidation policy.

ETag
string

Weak validator for the current model representation.

The current representation matches If-None-Match.

Cache-Control
string
Allowed value: no-cache, stale-if-error=300

Model catalog revalidation policy.

ETag
string

Weak validator for the current model representation.

An Authorization header was supplied but the SayGM API key is invalid.

Media typeapplication/json

OpenAI-shaped gateway error. Image errors after dispatch may also include billing fields at the top level.

object
error
required
object
message
required
string
type
required
string
code
required

Machine-readable error code. Account request limits use rate_limit_exceeded.

string
param
required
string | null
request_id

Gateway request identifier.

string
charged
boolean
billing_status
string
Allowed values: settled unbilled unknown
potentially_charged
boolean
cost_ndollars

Settled charge in nano-dollars; present when billing_status is settled.

integer
Example
{
"billing_status": "settled"
}
Cache-Control
string
Allowed value: no-store

The response is not stored.

No model with that id is in the catalog. The response carries error.code model_not_found.

Media typeapplication/json

OpenAI-shaped gateway error. Image errors after dispatch may also include billing fields at the top level.

object
error
required
object
message
required
string
type
required
string
code
required

Machine-readable error code. Account request limits use rate_limit_exceeded.

string
param
required
string | null
request_id

Gateway request identifier.

string
charged
boolean
billing_status
string
Allowed values: settled unbilled unknown
potentially_charged
boolean
cost_ndollars

Settled charge in nano-dollars; present when billing_status is settled.

integer
Example
{
"billing_status": "settled"
}
Cache-Control
string
Allowed value: no-store

The response is not stored.

When Authorization is present, the credit service is unavailable (ums_unavailable).

Media typeapplication/json

OpenAI-shaped gateway error. Image errors after dispatch may also include billing fields at the top level.

object
error
required
object
message
required
string
type
required
string
code
required

Machine-readable error code. Account request limits use rate_limit_exceeded.

string
param
required
string | null
request_id

Gateway request identifier.

string
charged
boolean
billing_status
string
Allowed values: settled unbilled unknown
potentially_charged
boolean
cost_ndollars

Settled charge in nano-dollars; present when billing_status is settled.

integer
Examples
ExamplecreditServiceUnavailable
{
"error": {
"message": "Credit service unavailable.",
"type": "server_error",
"code": "ums_unavailable",
"param": null
}
}
Cache-Control
string
Allowed value: no-store

The response is not stored.