From 5f553634394718a429e75b44c6cfec13b9c8d724 Mon Sep 17 00:00:00 2001 From: Guillaume Gay Date: Sun, 18 Oct 2026 05:39:41 +0000 Subject: [PATCH] =?UTF-8?q?chore(catalog):=20sync=20from=20gateway=20?= =?UTF-8?q?=E2=80=94=201=20added,=201=20corrected,=200=20retired?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Every added param and every corrected bound was verified end-to-end through the gateway. --- .../fireworks/deepseek-v4p1-flash-1017.yaml | 66 ++++++++++++++ models/xai/grok-4.6.yaml | 2 +- .../src/modelparams/_generated/catalog.json | 91 ++++++++++++++++++- .../src/modelparams/_generated/model_ids.py | 2 + .../src/modelparams/_generated/registry.py | 1 + .../src/modelparams/types/fireworks.py | 16 ++++ .../src/modelparams/types/xai.py | 2 +- packages/modelparams/src/generated/data.ts | 91 ++++++++++++++++++- .../modelparams/src/generated/defaults.ts | 1 + .../modelparams/src/generated/model-ids.ts | 1 + .../modelparams/src/generated/params-by-id.ts | 9 ++ 11 files changed, 278 insertions(+), 4 deletions(-) create mode 100644 models/fireworks/deepseek-v4p1-flash-1017.yaml diff --git a/models/fireworks/deepseek-v4p1-flash-1017.yaml b/models/fireworks/deepseek-v4p1-flash-1017.yaml new file mode 100644 index 00000000..2dce08e2 --- /dev/null +++ b/models/fireworks/deepseek-v4p1-flash-1017.yaml @@ -0,0 +1,66 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: fireworks +authType: api_key +apiSurface: openai-chat-completions +model: deepseek-v4p1-flash-1017 +wireId: accounts/fireworks/models/deepseek-v4p1-flash-1017 +params: + - path: max_tokens + type: integer + label: Max tokens + description: Maximum number of tokens to generate in the response. + range: + min: 1 + group: generation_length + - path: temperature + type: number + label: Temperature + description: Controls randomness. Lower values make outputs more focused; higher values make them more varied. + range: + min: 0 + max: 2 + step: 0.1 + group: sampling + - path: top_p + type: number + label: Top P + description: Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability. + range: + min: 0 + max: 1 + step: 0.01 + group: sampling + - path: top_k + type: integer + label: Top K + description: Limits token sampling to the top K most likely next tokens. + range: + min: 1 + max: 100 + group: sampling + - path: presence_penalty + type: number + label: Presence penalty + description: Penalizes tokens that have already appeared to encourage a wider variety of content. + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: frequency_penalty + type: number + label: Frequency penalty + description: Penalizes tokens in proportion to how often they have already appeared. + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: response_format.type + type: enum + label: Response format + description: Choose plain text or force valid JSON output. + values: + - text + - json_object + group: output_format diff --git a/models/xai/grok-4.6.yaml b/models/xai/grok-4.6.yaml index a24b7f2d..ab967076 100644 --- a/models/xai/grok-4.6.yaml +++ b/models/xai/grok-4.6.yaml @@ -28,7 +28,7 @@ params: description: Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability. default: 1 range: - min: 0 + min: 0.01 max: 1 step: 0.01 group: sampling diff --git a/packages/modelparams-python/src/modelparams/_generated/catalog.json b/packages/modelparams-python/src/modelparams/_generated/catalog.json index 9109f2e4..8a66feb2 100644 --- a/packages/modelparams-python/src/modelparams/_generated/catalog.json +++ b/packages/modelparams-python/src/modelparams/_generated/catalog.json @@ -15282,6 +15282,95 @@ } ] }, + { + "provider": "fireworks", + "authType": "api_key", + "apiSurface": "openai-chat-completions", + "model": "deepseek-v4p1-flash-1017", + "wireId": "accounts/fireworks/models/deepseek-v4p1-flash-1017", + "params": [ + { + "path": "max_tokens", + "label": "Max tokens", + "description": "Maximum number of tokens to generate in the response.", + "group": "generation_length", + "type": "integer", + "range": { + "min": 1 + } + }, + { + "path": "temperature", + "label": "Temperature", + "description": "Controls randomness. Lower values make outputs more focused; higher values make them more varied.", + "group": "sampling", + "type": "number", + "range": { + "min": 0, + "max": 2, + "step": 0.1 + } + }, + { + "path": "top_p", + "label": "Top P", + "description": "Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability.", + "group": "sampling", + "type": "number", + "range": { + "min": 0, + "max": 1, + "step": 0.01 + } + }, + { + "path": "top_k", + "label": "Top K", + "description": "Limits token sampling to the top K most likely next tokens.", + "group": "sampling", + "type": "integer", + "range": { + "min": 1, + "max": 100 + } + }, + { + "path": "presence_penalty", + "label": "Presence penalty", + "description": "Penalizes tokens that have already appeared to encourage a wider variety of content.", + "group": "sampling", + "type": "number", + "range": { + "min": -2, + "max": 2, + "step": 0.1 + } + }, + { + "path": "frequency_penalty", + "label": "Frequency penalty", + "description": "Penalizes tokens in proportion to how often they have already appeared.", + "group": "sampling", + "type": "number", + "range": { + "min": -2, + "max": 2, + "step": 0.1 + } + }, + { + "path": "response_format.type", + "label": "Response format", + "description": "Choose plain text or force valid JSON output.", + "group": "output_format", + "type": "enum", + "values": [ + "text", + "json_object" + ] + } + ] + }, { "provider": "fireworks", "authType": "api_key", @@ -35745,7 +35834,7 @@ "type": "number", "default": 1, "range": { - "min": 0, + "min": 0.01, "max": 1, "step": 0.01 } diff --git a/packages/modelparams-python/src/modelparams/_generated/model_ids.py b/packages/modelparams-python/src/modelparams/_generated/model_ids.py index ef44d636..f4ee624c 100644 --- a/packages/modelparams-python/src/modelparams/_generated/model_ids.py +++ b/packages/modelparams-python/src/modelparams/_generated/model_ids.py @@ -204,6 +204,7 @@ "fireworks/deepseek-v4-pro", "fireworks/deepseek-v4-pro-0813", "fireworks/deepseek-v4p1-flash", + "fireworks/deepseek-v4p1-flash-1017", "fireworks/ember-1", "fireworks/glm-5p2", "fireworks/glm-5p3", @@ -702,6 +703,7 @@ "fireworks/deepseek-v4-pro", "fireworks/deepseek-v4-pro-0813", "fireworks/deepseek-v4p1-flash", + "fireworks/deepseek-v4p1-flash-1017", "fireworks/ember-1", "fireworks/glm-5p2", "fireworks/glm-5p3", diff --git a/packages/modelparams-python/src/modelparams/_generated/registry.py b/packages/modelparams-python/src/modelparams/_generated/registry.py index e7d7465e..70759449 100644 --- a/packages/modelparams-python/src/modelparams/_generated/registry.py +++ b/packages/modelparams-python/src/modelparams/_generated/registry.py @@ -229,6 +229,7 @@ "fireworks/deepseek-v4-pro": fireworks.Deepseek_V4_ProParams, "fireworks/deepseek-v4-pro-0813": fireworks.Deepseek_V4_Pro_0813Params, "fireworks/deepseek-v4p1-flash": fireworks.Deepseek_V4p1_FlashParams, + "fireworks/deepseek-v4p1-flash-1017": fireworks.Deepseek_V4p1_Flash_1017Params, "fireworks/ember-1": fireworks.Ember_1Params, "fireworks/glm-5p2": fireworks.Glm_5p2Params, "fireworks/glm-5p3": fireworks.Glm_5p3Params, diff --git a/packages/modelparams-python/src/modelparams/types/fireworks.py b/packages/modelparams-python/src/modelparams/types/fireworks.py index e78e66d5..ac70a28f 100644 --- a/packages/modelparams-python/src/modelparams/types/fireworks.py +++ b/packages/modelparams-python/src/modelparams/types/fireworks.py @@ -85,6 +85,21 @@ ) setattr(Deepseek_V4p1_FlashParams, "__pydantic_config__", _PARAMS_CONFIG) +Deepseek_V4p1_Flash_1017Params = TypedDict( + "Deepseek_V4p1_Flash_1017Params", + { + "max_tokens": Annotated[int, Field(ge=1)], + "temperature": Annotated[float, Field(ge=0, le=2)], + "top_p": Annotated[float, Field(ge=0, le=1)], + "top_k": Annotated[int, Field(ge=1, le=100)], + "presence_penalty": Annotated[float, Field(ge=-2, le=2)], + "frequency_penalty": Annotated[float, Field(ge=-2, le=2)], + "response_format.type": Literal["text", "json_object"], + }, + total=False, +) +setattr(Deepseek_V4p1_Flash_1017Params, "__pydantic_config__", _PARAMS_CONFIG) + Ember_1Params = TypedDict( "Ember_1Params", { @@ -376,6 +391,7 @@ "Deepseek_V4_ProParams", "Deepseek_V4_Pro_0813Params", "Deepseek_V4p1_FlashParams", + "Deepseek_V4p1_Flash_1017Params", "Ember_1Params", "Glm_5p2Params", "Glm_5p3Params", diff --git a/packages/modelparams-python/src/modelparams/types/xai.py b/packages/modelparams-python/src/modelparams/types/xai.py index 4d9b70e3..fa88b167 100644 --- a/packages/modelparams-python/src/modelparams/types/xai.py +++ b/packages/modelparams-python/src/modelparams/types/xai.py @@ -138,7 +138,7 @@ { "max_completion_tokens": Annotated[int, Field(ge=1)], "temperature": Annotated[float, Field(ge=0, le=2)], - "top_p": Annotated[float, Field(ge=0, le=1)], + "top_p": Annotated[float, Field(ge=0.01, le=1)], "seed": int, "stop": str, "reasoning_effort": Literal["low", "medium", "high", "xhigh"], diff --git a/packages/modelparams/src/generated/data.ts b/packages/modelparams/src/generated/data.ts index 9afe26b1..204e4937 100644 --- a/packages/modelparams/src/generated/data.ts +++ b/packages/modelparams/src/generated/data.ts @@ -15287,6 +15287,95 @@ const GENERATED_CATALOG = [ } ] }, + { + "provider": "fireworks", + "authType": "api_key", + "apiSurface": "openai-chat-completions", + "model": "deepseek-v4p1-flash-1017", + "wireId": "accounts/fireworks/models/deepseek-v4p1-flash-1017", + "params": [ + { + "path": "max_tokens", + "label": "Max tokens", + "description": "Maximum number of tokens to generate in the response.", + "group": "generation_length", + "type": "integer", + "range": { + "min": 1 + } + }, + { + "path": "temperature", + "label": "Temperature", + "description": "Controls randomness. Lower values make outputs more focused; higher values make them more varied.", + "group": "sampling", + "type": "number", + "range": { + "min": 0, + "max": 2, + "step": 0.1 + } + }, + { + "path": "top_p", + "label": "Top P", + "description": "Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability.", + "group": "sampling", + "type": "number", + "range": { + "min": 0, + "max": 1, + "step": 0.01 + } + }, + { + "path": "top_k", + "label": "Top K", + "description": "Limits token sampling to the top K most likely next tokens.", + "group": "sampling", + "type": "integer", + "range": { + "min": 1, + "max": 100 + } + }, + { + "path": "presence_penalty", + "label": "Presence penalty", + "description": "Penalizes tokens that have already appeared to encourage a wider variety of content.", + "group": "sampling", + "type": "number", + "range": { + "min": -2, + "max": 2, + "step": 0.1 + } + }, + { + "path": "frequency_penalty", + "label": "Frequency penalty", + "description": "Penalizes tokens in proportion to how often they have already appeared.", + "group": "sampling", + "type": "number", + "range": { + "min": -2, + "max": 2, + "step": 0.1 + } + }, + { + "path": "response_format.type", + "label": "Response format", + "description": "Choose plain text or force valid JSON output.", + "group": "output_format", + "type": "enum", + "values": [ + "text", + "json_object" + ] + } + ] + }, { "provider": "fireworks", "authType": "api_key", @@ -35750,7 +35839,7 @@ const GENERATED_CATALOG = [ "type": "number", "default": 1, "range": { - "min": 0, + "min": 0.01, "max": 1, "step": 0.01 } diff --git a/packages/modelparams/src/generated/defaults.ts b/packages/modelparams/src/generated/defaults.ts index e991df37..6d7522f5 100644 --- a/packages/modelparams/src/generated/defaults.ts +++ b/packages/modelparams/src/generated/defaults.ts @@ -848,6 +848,7 @@ export const DEFAULTS = { "fireworks/deepseek-v4-pro": {}, "fireworks/deepseek-v4-pro-0813": {}, "fireworks/deepseek-v4p1-flash": {}, + "fireworks/deepseek-v4p1-flash-1017": {}, "fireworks/ember-1": {}, "fireworks/glm-5p2": {}, "fireworks/glm-5p3": {}, diff --git a/packages/modelparams/src/generated/model-ids.ts b/packages/modelparams/src/generated/model-ids.ts index 9c8b5830..3dec8824 100644 --- a/packages/modelparams/src/generated/model-ids.ts +++ b/packages/modelparams/src/generated/model-ids.ts @@ -202,6 +202,7 @@ export const MODEL_IDS = [ "fireworks/deepseek-v4-pro", "fireworks/deepseek-v4-pro-0813", "fireworks/deepseek-v4p1-flash", + "fireworks/deepseek-v4p1-flash-1017", "fireworks/ember-1", "fireworks/glm-5p2", "fireworks/glm-5p3", diff --git a/packages/modelparams/src/generated/params-by-id.ts b/packages/modelparams/src/generated/params-by-id.ts index bce7c58f..4003a27c 100644 --- a/packages/modelparams/src/generated/params-by-id.ts +++ b/packages/modelparams/src/generated/params-by-id.ts @@ -1495,6 +1495,15 @@ export type ParamsById = { frequency_penalty: number; "response_format.type": "text" | "json_object"; }; + "fireworks/deepseek-v4p1-flash-1017": { + max_tokens: number; + temperature: number; + top_p: number; + top_k: number; + presence_penalty: number; + frequency_penalty: number; + "response_format.type": "text" | "json_object"; + }; "fireworks/ember-1": { max_tokens: number; temperature: number;