modelparams.dev
API GitHub

Providers

Alibaba Anthropic Cerebras Cohere DeepSeek Google Groq Meta MiniMax Mistral Moonshot AI NVIDIA OpenAI Perplexity Thinking Machines xAI Xiaomi Z.ai

Providers

Alibaba Anthropic Cerebras Cohere DeepSeek Google Groq Meta MiniMax Mistral Moonshot AI NVIDIA OpenAI Perplexity Thinking Machines xAI Xiaomi Z.ai
OpenAI 6 params

OpenAI GPT-OSS 20B API parameters

These are the API parameters modelparams.dev tracks for OpenAI GPT-OSS 20B — the settings you send in a request. Each row gives the type, default, valid range or values, and the conditions that gate it. It's the same data the JSON API serves.

After the parameter count instead — how many weights GPT-OSS 20B has? That's a different number, and we don't track it. Here's the difference.

Length 1 param
Parameter Type Default Description Condition
Max completion tokens
max_completion_tokens 		integer (1…131072) — Maximum number of output tokens the model may generate. —
Sampling 2 params
Parameter Type Default Description Condition
Temperature
temperature 		number 1 Controls randomness. Lower values make outputs more focused; higher values make them more varied. OpenAI recommends sampling at 1.0. —
Top P
top_p 		number (0…1) 1 Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability. OpenAI recommends sampling at 1.0. —
Reasoning 1 param
Parameter Type Default Description Condition
Reasoning effort
reasoning_effort 		enum (low | medium | high) "medium" Controls how much reasoning the model should perform before producing an answer. —
Tools 1 param
Parameter Type Default Description Condition
Tool choice
tool_choice 		enum (auto | none | required) — Controls whether the model may call tools, must call one, or skips tool calls. —
Output 1 param
Parameter Type Default Description Condition
Response format
response_format.type 		enum (text | json_schema) "text" Controls whether the model returns normal text or a schema-constrained JSON object. —

OpenAI GPT-OSS 20B API parameters in brief

OpenAI GPT-OSS 20B documents 6 API parameters, grouped by what they control:

Frequently asked questions

Which API parameters does OpenAI GPT-OSS 20B support?
OpenAI GPT-OSS 20B accepts 6 API parameters in the request body: max_completion_tokens, temperature, top_p, reasoning_effort, response_format.type, tool_choice.
What is the default temperature for OpenAI GPT-OSS 20B?
The default temperature for OpenAI GPT-OSS 20B is 1.
What is the default top_p for OpenAI GPT-OSS 20B?
The default top_p for OpenAI GPT-OSS 20B is 1, within a valid range of 0 to 1.

Resources

All OpenAI models Glossary Full catalog

GPT-OSS 20B — JSON

The full model definition as served by the API. Copy it or open the endpoint directly.

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

Other OpenAI models

ChatGPT 4o Latest 3 params View GPT-3.5 Turbo 3 params View GPT-4 Turbo 3 params View GPT-4 Turbo 2024-04-09 3 params View GPT-4.1 3 params View GPT-4.1 Mini 3 params View GPT-4.1 Nano 3 params View GPT-4o 3 params View GPT-4o 2024-11-20 3 params View GPT-4o mini 3 params View GPT-5 2 params View GPT-5 Chat Latest 1 param View GPT-5 Mini 2 params View GPT-5 Nano 2 params View GPT-5.1 2 params View GPT-5.1 Codex Subscription 3 params View GPT-5.1 Codex Max Subscription 3 params View GPT-5.2 2 params View GPT-5.2 Subscription 3 params View GPT-5.2 Codex Subscription 3 params View GPT-5.3 Codex 2 params View GPT-5.3 Codex Subscription 3 params View GPT-5.3 Codex Spark Subscription 3 params View GPT-5.4 2 params View GPT-5.4 Subscription 3 params View GPT-5.4 Mini 2 params View GPT-5.4 Mini Subscription 3 params View GPT-5.4 Nano 2 params View GPT-5.4 Pro 2 params View GPT-5.4 Pro Subscription 3 params View GPT-5.5 2 params View GPT-5.5 Subscription 3 params View GPT-5.5 Pro 2 params View GPT-5.5 Pro Subscription 3 params View GPT-5.6 Luna 2 params View GPT-5.6 Luna Subscription 3 params View GPT-5.6 Sol 2 params View GPT-5.6 Sol Subscription 3 params View GPT-5.6 Terra 2 params View GPT-5.6 Terra Subscription 3 params View GPT-OSS 120B 6 params View GPT-OSS Safeguard 120B 4 params View GPT-OSS Safeguard 20B 4 params View o1 2 params View o1-mini 2 params View o1-preview 2 params View o3 2 params View o3-mini 2 params View o3-pro 2 params View o4-mini 2 params View

How to use

Building with an AI agent? Hit Copy to grab this whole guide as Markdown and paste it in — or point your agent straight at /llms.txt.

modelparams.dev is an open, community-maintained catalog of model parameters. Each entry shows the knobs you can turn — type, default, range, and the conditions that gate it.

The same model accessed via an API key and via a subscription usually exposes a different set of parameters. We list both as separate entries so the data stays honest.

Catalog API

The full catalog is static JSON, CORS-enabled, served from the edge.

curl https://modelparams.dev/api/v1/models.json

Each entry is keyed by provider/model for API-key variants; subscription variants append -subscription.

If you only need the params for one model contract, use the providerless endpoint. Subscription contracts are model slugs with -subscription. 

curl https://modelparams.dev/api/v1/params/gpt-5.5.json
curl https://modelparams.dev/api/v1/params/gpt-5.5-subscription.json

Single model

curl https://modelparams.dev/api/v1/models/anthropic/claude-opus-4-7.json
curl https://modelparams.dev/api/v1/models/anthropic/claude-opus-4-7-subscription.json

JSON Schema

Every entry validates against a JSON Schema you can use in your editor or pipeline. 

curl https://modelparams.dev/api/v1/schema.json

Add this header to any YAML you author for autocomplete in VS Code: 

# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json

Logos

Provider logos are available at /assets/logos/{provider}.svg where {provider} is the provider slug. They use currentColor so they inherit your text color. 

curl https://modelparams.dev/assets/logos/anthropic.svg

Logos are sourced from the models.dev repo (MIT) and used under nominative fair use.

Contribute

The data lives in YAML under models/{provider}/{model}-{auth}.yaml in the GitHub repo. Open a PR; CI validates against the schema and rebuilds.

Edit on GitHub MIT licensed