feat: add gpt-oss family + Thinking Machines Inkling

models/google/gemma-4-26b-a4b-it.yaml

@@ -50,6 +50,14 @@ params:
     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.thinkingLevel
+    type: enum
+    label: Thinking level
+    description: Toggles Gemma 4 reasoning; high enables thinking and minimal disables it.
+    values:
+      - minimal
+      - high
+    group: reasoning
   - path: generationConfig.thinkingConfig.includeThoughts
     type: boolean
     label: Include thoughts

models/google/gemma-4-31b-it.yaml

@@ -50,6 +50,14 @@ params:
     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.thinkingLevel
+    type: enum
+    label: Thinking level
+    description: Toggles Gemma 4 reasoning; high enables thinking and minimal disables it.
+    values:
+      - minimal
+      - high
+    group: reasoning
   - path: generationConfig.thinkingConfig.includeThoughts
     type: boolean
     label: Include thoughts

models/openai/gpt-oss-120b.yaml

@@ -0,0 +1,56 @@
+# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json
+provider: openai
+authType: api_key
+model: gpt-oss-120b
+params:
+  - path: max_completion_tokens
+    type: integer
+    label: Max tokens
+    description: Maximum number of output tokens the model may generate.
+    range:
+      min: 1
+      max: 131072
+    group: generation_length
+  - path: temperature
+    type: number
+    label: Temperature
+    description: Controls randomness. Lower values make outputs more focused; higher values make them more varied. OpenAI recommends sampling at 1.0.
+    default: 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. OpenAI recommends sampling at 1.0.
+    default: 1
+    range:
+      min: 0
+      max: 1
+    group: sampling
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much reasoning the model should perform before producing an answer.
+    default: medium
+    values:
+      - low
+      - medium
+      - high
+    group: reasoning
+  - path: response_format.type
+    type: enum
+    label: Response format
+    description: Controls whether the model returns normal text or a schema-constrained JSON object.
+    default: text
+    values:
+      - text
+      - json_schema
+    group: output_format
+  - path: tool_choice
+    type: enum
+    label: Tool choice
+    description: Controls whether the model may call tools, must call one, or skips tool calls.
+    values:
+      - auto
+      - none
+      - required
+    group: tooling

models/openai/gpt-oss-20b.yaml

@@ -0,0 +1,56 @@
+# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json
+provider: openai
+authType: api_key
+model: gpt-oss-20b
+params:
+  - path: max_completion_tokens
+    type: integer
+    label: Max tokens
+    description: Maximum number of output tokens the model may generate.
+    range:
+      min: 1
+      max: 131072
+    group: generation_length
+  - path: temperature
+    type: number
+    label: Temperature
+    description: Controls randomness. Lower values make outputs more focused; higher values make them more varied. OpenAI recommends sampling at 1.0.
+    default: 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. OpenAI recommends sampling at 1.0.
+    default: 1
+    range:
+      min: 0
+      max: 1
+    group: sampling
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much reasoning the model should perform before producing an answer.
+    default: medium
+    values:
+      - low
+      - medium
+      - high
+    group: reasoning
+  - path: response_format.type
+    type: enum
+    label: Response format
+    description: Controls whether the model returns normal text or a schema-constrained JSON object.
+    default: text
+    values:
+      - text
+      - json_schema
+    group: output_format
+  - path: tool_choice
+    type: enum
+    label: Tool choice
+    description: Controls whether the model may call tools, must call one, or skips tool calls.
+    values:
+      - auto
+      - none
+      - required
+    group: tooling

models/openai/gpt-oss-safeguard-120b.yaml

@@ -0,0 +1,38 @@
+# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json
+provider: openai
+authType: api_key
+model: gpt-oss-safeguard-120b
+params:
+  - path: max_completion_tokens
+    type: integer
+    label: Max tokens
+    description: Maximum number of output tokens the model may generate.
+    range:
+      min: 1
+      max: 131072
+    group: generation_length
+  - path: temperature
+    type: number
+    label: Temperature
+    description: Controls randomness. Lower values make outputs more focused; higher values make them more varied. OpenAI recommends sampling at 1.0.
+    default: 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. OpenAI recommends sampling at 1.0.
+    default: 1
+    range:
+      min: 0
+      max: 1
+    group: sampling
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much reasoning the model performs when interpreting the provided safety policy before returning a classification.
+    default: medium
+    values:
+      - low
+      - medium
+      - high
+    group: reasoning

models/openai/gpt-oss-safeguard-20b.yaml

@@ -0,0 +1,38 @@
+# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json
+provider: openai
+authType: api_key
+model: gpt-oss-safeguard-20b
+params:
+  - path: max_completion_tokens
+    type: integer
+    label: Max tokens
+    description: Maximum number of output tokens the model may generate.
+    range:
+      min: 1
+      max: 131072
+    group: generation_length
+  - path: temperature
+    type: number
+    label: Temperature
+    description: Controls randomness. Lower values make outputs more focused; higher values make them more varied. OpenAI recommends sampling at 1.0.
+    default: 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. OpenAI recommends sampling at 1.0.
+    default: 1
+    range:
+      min: 0
+      max: 1
+    group: sampling
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much reasoning the model performs when interpreting the provided safety policy before returning a classification.
+    default: medium
+    values:
+      - low
+      - medium
+      - high
+    group: reasoning

models/thinking-machines/Inkling.yaml

@@ -0,0 +1,46 @@
+# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json
+provider: thinking-machines
+authType: api_key
+model: Inkling
+params:
+  - path: max_tokens
+    type: integer
+    label: Max tokens
+    description: Maximum number of output tokens the model may generate.
+    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: 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
+    group: sampling
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much thinking Inkling performs before answering. Accepts a preset name or a number between 0.0 and 0.99; presets map to numeric effort levels (none=0.0, minimal=0.1, low=0.2, medium=0.7, high=0.9, xhigh=0.99).
+    default: high
+    values:
+      - none
+      - minimal
+      - low
+      - medium
+      - high
+      - xhigh
+    group: reasoning
+  - path: separate_reasoning
+    type: boolean
+    label: Separate reasoning
+    description: Returns the model's reasoning in a dedicated reasoning_content field instead of interleaving it with the final message content.
+    default: true
+    group: reasoning