diff --git a/models/google/gemini-4-argon.yaml b/models/google/gemini-4-argon.yaml new file mode 100644 index 00000000..ee25f9e6 --- /dev/null +++ b/models/google/gemini-4-argon.yaml @@ -0,0 +1,65 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: google +authType: api_key +apiSurface: google-generate-content +model: gemini-4-argon +status: active +replacement: google/gemini-3.1-flash-lite +params: + - path: generationConfig.maxOutputTokens + type: integer + label: Max output tokens + description: Maximum number of tokens to include in a response candidate. + range: + min: 1 + max: 65536 + group: generation_length + - path: generationConfig.temperature + type: number + label: Temperature + description: Controls randomness. Lower values make outputs more focused; higher values make them more varied. + default: 1 + range: + min: 0 + max: 2 + step: 0.1 + group: sampling + - path: generationConfig.topP + type: number + label: Top P + description: Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability. + default: 0.95 + range: + min: 0 + max: 1 + step: 0.01 + group: sampling + - path: generationConfig.topK + type: integer + label: Top K + description: Limits token sampling to the top K most likely next tokens. + default: 64 + range: + min: 0 + group: sampling + - path: generationConfig.thinkingConfig.thinkingBudget + type: integer + label: Thinking budget + description: Number of thinking tokens Gemini should use; -1 uses dynamic thinking, 0 disables thinking, and fixed budgets start at 512 tokens. + default: 0 + group: reasoning + - path: generationConfig.thinkingConfig.includeThoughts + type: boolean + label: Include thoughts + description: Controls whether Gemini returns available thought summaries in the response parts. + default: false + group: reasoning + - path: generationConfig.responseMimeType + type: enum + label: Response MIME type + description: MIME type for generated text candidates. + default: text/plain + values: + - text/plain + - application/json + group: output_format diff --git a/packages/modelparams-python/src/modelparams/_generated/catalog.json b/packages/modelparams-python/src/modelparams/_generated/catalog.json index 9ac5f3d2..f84cf7c8 100644 --- a/packages/modelparams-python/src/modelparams/_generated/catalog.json +++ b/packages/modelparams-python/src/modelparams/_generated/catalog.json @@ -18728,6 +18728,92 @@ } ] }, + { + "provider": "google", + "authType": "api_key", + "apiSurface": "google-generate-content", + "model": "gemini-4-argon", + "status": "active", + "replacement": "google/gemini-3.1-flash-lite", + "params": [ + { + "path": "generationConfig.maxOutputTokens", + "label": "Max output tokens", + "description": "Maximum number of tokens to include in a response candidate.", + "group": "generation_length", + "type": "integer", + "range": { + "min": 1, + "max": 65536 + } + }, + { + "path": "generationConfig.temperature", + "label": "Temperature", + "description": "Controls randomness. Lower values make outputs more focused; higher values make them more varied.", + "group": "sampling", + "type": "number", + "default": 1, + "range": { + "min": 0, + "max": 2, + "step": 0.1 + } + }, + { + "path": "generationConfig.topP", + "label": "Top P", + "description": "Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability.", + "group": "sampling", + "type": "number", + "default": 0.95, + "range": { + "min": 0, + "max": 1, + "step": 0.01 + } + }, + { + "path": "generationConfig.topK", + "label": "Top K", + "description": "Limits token sampling to the top K most likely next tokens.", + "group": "sampling", + "type": "integer", + "default": 64, + "range": { + "min": 0 + } + }, + { + "path": "generationConfig.thinkingConfig.thinkingBudget", + "label": "Thinking budget", + "description": "Number of thinking tokens Gemini should use; -1 uses dynamic thinking, 0 disables thinking, and fixed budgets start at 512 tokens.", + "group": "reasoning", + "type": "integer", + "default": 0 + }, + { + "path": "generationConfig.thinkingConfig.includeThoughts", + "label": "Include thoughts", + "description": "Controls whether Gemini returns available thought summaries in the response parts.", + "group": "reasoning", + "type": "boolean", + "default": false + }, + { + "path": "generationConfig.responseMimeType", + "label": "Response MIME type", + "description": "MIME type for generated text candidates.", + "group": "output_format", + "type": "enum", + "default": "text/plain", + "values": [ + "text/plain", + "application/json" + ] + } + ] + }, { "provider": "google", "authType": "api_key", diff --git a/packages/modelparams-python/src/modelparams/_generated/model_ids.py b/packages/modelparams-python/src/modelparams/_generated/model_ids.py index babf520d..970579ff 100644 --- a/packages/modelparams-python/src/modelparams/_generated/model_ids.py +++ b/packages/modelparams-python/src/modelparams/_generated/model_ids.py @@ -241,6 +241,7 @@ "google/gemini-3.6-flash", "google/gemini-3.7-flash", "google/gemini-3.8-flash", + "google/gemini-4-argon", "google/gemini-flash-latest", "google/gemma-3-12b-it", "google/gemma-3-1b-it", @@ -734,6 +735,7 @@ "google/gemini-3.6-flash", "google/gemini-3.7-flash", "google/gemini-3.8-flash", + "google/gemini-4-argon", "google/gemini-flash-latest", "google/gemma-3-12b-it", "google/gemma-3-1b-it", diff --git a/packages/modelparams-python/src/modelparams/_generated/registry.py b/packages/modelparams-python/src/modelparams/_generated/registry.py index 09a51759..ede7842c 100644 --- a/packages/modelparams-python/src/modelparams/_generated/registry.py +++ b/packages/modelparams-python/src/modelparams/_generated/registry.py @@ -266,6 +266,7 @@ "google/gemini-3.6-flash": google.Gemini_3_6_FlashParams, "google/gemini-3.7-flash": google.Gemini_3_7_FlashParams, "google/gemini-3.8-flash": google.Gemini_3_8_FlashParams, + "google/gemini-4-argon": google.Gemini_4_ArgonParams, "google/gemini-flash-latest": google.Gemini_Flash_LatestParams, "google/gemma-3-12b-it": google.Gemma_3_12b_ItParams, "google/gemma-3-1b-it": google.Gemma_3_1b_ItParams, diff --git a/packages/modelparams-python/src/modelparams/types/google.py b/packages/modelparams-python/src/modelparams/types/google.py index 57a2e0ee..750391ec 100644 --- a/packages/modelparams-python/src/modelparams/types/google.py +++ b/packages/modelparams-python/src/modelparams/types/google.py @@ -330,6 +330,21 @@ ) setattr(Gemini_3_8_FlashParams, "__pydantic_config__", _PARAMS_CONFIG) +Gemini_4_ArgonParams = TypedDict( + "Gemini_4_ArgonParams", + { + "generationConfig.maxOutputTokens": Annotated[int, Field(ge=1, le=65536)], + "generationConfig.temperature": Annotated[float, Field(ge=0, le=2)], + "generationConfig.topP": Annotated[float, Field(ge=0, le=1)], + "generationConfig.topK": Annotated[int, Field(ge=0)], + "generationConfig.thinkingConfig.thinkingBudget": int, + "generationConfig.thinkingConfig.includeThoughts": bool, + "generationConfig.responseMimeType": Literal["text/plain", "application/json"], + }, + total=False, +) +setattr(Gemini_4_ArgonParams, "__pydantic_config__", _PARAMS_CONFIG) + Gemini_Flash_LatestParams = TypedDict( "Gemini_Flash_LatestParams", { @@ -509,6 +524,7 @@ "Gemini_3_6_FlashParams", "Gemini_3_7_FlashParams", "Gemini_3_8_FlashParams", + "Gemini_4_ArgonParams", "Gemini_Flash_LatestParams", "Gemma_3_12b_ItParams", "Gemma_3_1b_ItParams", diff --git a/packages/modelparams/src/generated/data.ts b/packages/modelparams/src/generated/data.ts index 9b61ab0b..2cd83d3d 100644 --- a/packages/modelparams/src/generated/data.ts +++ b/packages/modelparams/src/generated/data.ts @@ -18733,6 +18733,92 @@ const GENERATED_CATALOG = [ } ] }, + { + "provider": "google", + "authType": "api_key", + "apiSurface": "google-generate-content", + "model": "gemini-4-argon", + "status": "active", + "replacement": "google/gemini-3.1-flash-lite", + "params": [ + { + "path": "generationConfig.maxOutputTokens", + "label": "Max output tokens", + "description": "Maximum number of tokens to include in a response candidate.", + "group": "generation_length", + "type": "integer", + "range": { + "min": 1, + "max": 65536 + } + }, + { + "path": "generationConfig.temperature", + "label": "Temperature", + "description": "Controls randomness. Lower values make outputs more focused; higher values make them more varied.", + "group": "sampling", + "type": "number", + "default": 1, + "range": { + "min": 0, + "max": 2, + "step": 0.1 + } + }, + { + "path": "generationConfig.topP", + "label": "Top P", + "description": "Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability.", + "group": "sampling", + "type": "number", + "default": 0.95, + "range": { + "min": 0, + "max": 1, + "step": 0.01 + } + }, + { + "path": "generationConfig.topK", + "label": "Top K", + "description": "Limits token sampling to the top K most likely next tokens.", + "group": "sampling", + "type": "integer", + "default": 64, + "range": { + "min": 0 + } + }, + { + "path": "generationConfig.thinkingConfig.thinkingBudget", + "label": "Thinking budget", + "description": "Number of thinking tokens Gemini should use; -1 uses dynamic thinking, 0 disables thinking, and fixed budgets start at 512 tokens.", + "group": "reasoning", + "type": "integer", + "default": 0 + }, + { + "path": "generationConfig.thinkingConfig.includeThoughts", + "label": "Include thoughts", + "description": "Controls whether Gemini returns available thought summaries in the response parts.", + "group": "reasoning", + "type": "boolean", + "default": false + }, + { + "path": "generationConfig.responseMimeType", + "label": "Response MIME type", + "description": "MIME type for generated text candidates.", + "group": "output_format", + "type": "enum", + "default": "text/plain", + "values": [ + "text/plain", + "application/json" + ] + } + ] + }, { "provider": "google", "authType": "api_key", diff --git a/packages/modelparams/src/generated/defaults.ts b/packages/modelparams/src/generated/defaults.ts index 13d20dae..2e9ba26c 100644 --- a/packages/modelparams/src/generated/defaults.ts +++ b/packages/modelparams/src/generated/defaults.ts @@ -1020,6 +1020,14 @@ export const DEFAULTS = { "generationConfig.thinkingConfig.includeThoughts": false, "generationConfig.responseMimeType": "text/plain", }, + "google/gemini-4-argon": { + "generationConfig.temperature": 1, + "generationConfig.topP": 0.95, + "generationConfig.topK": 64, + "generationConfig.thinkingConfig.thinkingBudget": 0, + "generationConfig.thinkingConfig.includeThoughts": false, + "generationConfig.responseMimeType": "text/plain", + }, "google/gemini-flash-latest": { "generationConfig.temperature": 1, "generationConfig.topP": 0.95, diff --git a/packages/modelparams/src/generated/model-ids.ts b/packages/modelparams/src/generated/model-ids.ts index 49023e98..f901e27e 100644 --- a/packages/modelparams/src/generated/model-ids.ts +++ b/packages/modelparams/src/generated/model-ids.ts @@ -239,6 +239,7 @@ export const MODEL_IDS = [ "google/gemini-3.6-flash", "google/gemini-3.7-flash", "google/gemini-3.8-flash", + "google/gemini-4-argon", "google/gemini-flash-latest", "google/gemma-3-12b-it", "google/gemma-3-1b-it", diff --git a/packages/modelparams/src/generated/params-by-id.ts b/packages/modelparams/src/generated/params-by-id.ts index 7d85caa2..0e3faaa9 100644 --- a/packages/modelparams/src/generated/params-by-id.ts +++ b/packages/modelparams/src/generated/params-by-id.ts @@ -1849,6 +1849,15 @@ export type ParamsById = { "generationConfig.thinkingConfig.includeThoughts": boolean; "generationConfig.responseMimeType": "text/plain" | "application/json"; }; + "google/gemini-4-argon": { + "generationConfig.maxOutputTokens": number; + "generationConfig.temperature": number; + "generationConfig.topP": number; + "generationConfig.topK": number; + "generationConfig.thinkingConfig.thinkingBudget": number; + "generationConfig.thinkingConfig.includeThoughts": boolean; + "generationConfig.responseMimeType": "text/plain" | "application/json"; + }; "google/gemini-flash-latest": { "generationConfig.maxOutputTokens": number; "generationConfig.temperature": number;