diff --git a/models/moonshot/kimi-k2.5.yaml b/models/moonshot/kimi-k2.5.yaml new file mode 100644 index 0000000..9bb3ada --- /dev/null +++ b/models/moonshot/kimi-k2.5.yaml @@ -0,0 +1,31 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: moonshot +authType: api_key +model: kimi-k2.5 +params: + - path: max_completion_tokens + type: integer + label: Max tokens + description: Maximum number of tokens to generate in the chat completion. + range: + min: 1 + group: generation_length + - path: thinking.type + type: enum + label: Thinking mode + description: >- + Controls whether Kimi reasons step by step before answering, or responds directly when set to + disabled. + values: + - enabled + - disabled + group: reasoning + - path: response_format.type + type: enum + label: Response format + description: Forces the response into plain text or a JSON object. + default: text + values: + - text + - json_object + group: output_format diff --git a/models/moonshot/kimi-k2.6.yaml b/models/moonshot/kimi-k2.6.yaml new file mode 100644 index 0000000..9bb30a1 --- /dev/null +++ b/models/moonshot/kimi-k2.6.yaml @@ -0,0 +1,32 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: moonshot +authType: api_key +model: kimi-k2.6 +params: + - path: max_completion_tokens + type: integer + label: Max tokens + description: Maximum number of tokens to generate in the chat completion. + range: + min: 1 + group: generation_length + - path: thinking.type + type: enum + label: Thinking mode + description: >- + Controls whether Kimi reasons step by step before answering. Thinking is enabled by default; + set disabled to respond directly. + default: enabled + values: + - enabled + - disabled + group: reasoning + - path: response_format.type + type: enum + label: Response format + description: Forces the response into plain text or a JSON object. + default: text + values: + - text + - json_object + group: output_format diff --git a/models/moonshot/moonshot-v1-128k.yaml b/models/moonshot/moonshot-v1-128k.yaml new file mode 100644 index 0000000..f9290cf --- /dev/null +++ b/models/moonshot/moonshot-v1-128k.yaml @@ -0,0 +1,75 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: moonshot +authType: api_key +model: moonshot-v1-128k +params: + - path: max_completion_tokens + type: integer + label: Max tokens + description: Maximum number of tokens to generate in the chat completion. + 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. + default: 0.3 + range: + min: 0 + max: 1 + 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. + default: 1 + range: + min: 0 + max: 1 + step: 0.01 + group: sampling + - path: n + type: integer + label: Number of completions + description: How many chat completion choices to generate for the request. + default: 1 + range: + min: 1 + max: 5 + group: generation_length + - path: presence_penalty + type: number + label: Presence penalty + description: >- + Penalizes tokens that have already appeared, encouraging the model to talk about new topics. + default: 0 + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: frequency_penalty + type: number + label: Frequency penalty + description: Penalizes tokens by how often they have appeared, reducing verbatim repetition. + default: 0 + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: response_format.type + type: enum + label: Response format + description: Forces the response into plain text or a JSON object. + default: text + values: + - text + - json_object + group: output_format diff --git a/models/moonshot/moonshot-v1-32k.yaml b/models/moonshot/moonshot-v1-32k.yaml new file mode 100644 index 0000000..a0c3df4 --- /dev/null +++ b/models/moonshot/moonshot-v1-32k.yaml @@ -0,0 +1,75 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: moonshot +authType: api_key +model: moonshot-v1-32k +params: + - path: max_completion_tokens + type: integer + label: Max tokens + description: Maximum number of tokens to generate in the chat completion. + 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. + default: 0.3 + range: + min: 0 + max: 1 + 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. + default: 1 + range: + min: 0 + max: 1 + step: 0.01 + group: sampling + - path: n + type: integer + label: Number of completions + description: How many chat completion choices to generate for the request. + default: 1 + range: + min: 1 + max: 5 + group: generation_length + - path: presence_penalty + type: number + label: Presence penalty + description: >- + Penalizes tokens that have already appeared, encouraging the model to talk about new topics. + default: 0 + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: frequency_penalty + type: number + label: Frequency penalty + description: Penalizes tokens by how often they have appeared, reducing verbatim repetition. + default: 0 + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: response_format.type + type: enum + label: Response format + description: Forces the response into plain text or a JSON object. + default: text + values: + - text + - json_object + group: output_format diff --git a/models/moonshot/moonshot-v1-8k.yaml b/models/moonshot/moonshot-v1-8k.yaml new file mode 100644 index 0000000..3d09595 --- /dev/null +++ b/models/moonshot/moonshot-v1-8k.yaml @@ -0,0 +1,75 @@ +# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json +provider: moonshot +authType: api_key +model: moonshot-v1-8k +params: + - path: max_completion_tokens + type: integer + label: Max tokens + description: Maximum number of tokens to generate in the chat completion. + 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. + default: 0.3 + range: + min: 0 + max: 1 + 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. + default: 1 + range: + min: 0 + max: 1 + step: 0.01 + group: sampling + - path: n + type: integer + label: Number of completions + description: How many chat completion choices to generate for the request. + default: 1 + range: + min: 1 + max: 5 + group: generation_length + - path: presence_penalty + type: number + label: Presence penalty + description: >- + Penalizes tokens that have already appeared, encouraging the model to talk about new topics. + default: 0 + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: frequency_penalty + type: number + label: Frequency penalty + description: Penalizes tokens by how often they have appeared, reducing verbatim repetition. + default: 0 + range: + min: -2 + max: 2 + step: 0.1 + group: sampling + - path: response_format.type + type: enum + label: Response format + description: Forces the response into plain text or a JSON object. + default: text + values: + - text + - json_object + group: output_format diff --git a/src/client/logos/moonshot.svg b/src/client/logos/moonshot.svg new file mode 100644 index 0000000..3cdf7c8 --- /dev/null +++ b/src/client/logos/moonshot.svg @@ -0,0 +1,3 @@ + + + diff --git a/src/data/display.ts b/src/data/display.ts index 1735797..703bfb6 100644 --- a/src/data/display.ts +++ b/src/data/display.ts @@ -11,6 +11,7 @@ const PROVIDER_LABELS: Record = { minimax: "MiniMax", cohere: "Cohere", perplexity: "Perplexity", + moonshot: "Moonshot AI", }; const MODEL_LABEL_OVERRIDES: Record = { @@ -28,6 +29,11 @@ const MODEL_LABEL_OVERRIDES: Record = { "minimax/minimax-m2.5-highspeed": "MiniMax M2.5 Highspeed", "minimax/minimax-m2.7": "MiniMax M2.7", "minimax/minimax-m2.7-highspeed": "MiniMax M2.7 Highspeed", + "moonshot/kimi-k2.6": "Kimi K2.6", + "moonshot/kimi-k2.5": "Kimi K2.5", + "moonshot/moonshot-v1-8k": "Moonshot v1 8K", + "moonshot/moonshot-v1-32k": "Moonshot v1 32K", + "moonshot/moonshot-v1-128k": "Moonshot v1 128K", }; const AUTH_LABELS: Record = {