modelparams.dev
API MCP GitHub

Providers

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

Providers

Alibaba Anthropic Bedrock Cerebras Cohere DeepSeek Fireworks Google Groq Meta MiniMax Mistral Moonshot AI NVIDIA OpenAI Opencode Go Perplexity Thinking Machines Vertex xAI Xiaomi Z.ai
NVIDIA Chat Completions 7 params

NVIDIA Llama 3.3 Nemotron Super 49B v1 API parameters

These are the API parameters modelparams.dev tracks for NVIDIA Llama 3.3 Nemotron Super 49B v1 on Chat Completions with an API key — 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 Llama 3.3 Nemotron Super 49B v1 has? That's a different number, and we don't track it. Here's the difference.

Length 2 params
Parameter Type Default Description Condition
Max tokens
max_tokens 		integer (1…16384) 4096 Maximum number of tokens to generate. Generation stops when this limit is reached. —
Stop
stop 		string — A string or list of strings where the API will stop generating further tokens. The returned text will not contain the stop sequence. —
Sampling 5 params
Parameter Type Default Description Condition
Temperature
temperature 		number (0…1) 0.6 Controls randomness. Lower values make outputs more focused; higher values make them more varied. Not recommended to modify both temperature and top_p in the same call. —
Top P
top_p 		number (-∞…1) 0.95 Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability. Not recommended to modify both temperature and top_p in the same call. —
Frequency penalty
frequency_penalty 		number (-2…2) 0 Penalizes new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. —
Presence penalty
presence_penalty 		number (-2…2) 0 Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. —
Seed
seed 		integer (0…18446744073709552000) 0 Best-effort deterministic sampling seed. Changing the seed produces a different response with similar characteristics. Fix the seed to reproduce results. —

NVIDIA Llama 3.3 Nemotron Super 49B v1 API parameters in brief

NVIDIA Llama 3.3 Nemotron Super 49B v1 on Chat Completions documents 7 API parameters, grouped by what they control:

Frequently asked questions

Which API parameters does NVIDIA Llama 3.3 Nemotron Super 49B v1 support?
NVIDIA Llama 3.3 Nemotron Super 49B v1 accepts 7 API parameters in the request body: temperature, top_p, max_tokens, frequency_penalty, presence_penalty, seed, and more.
What is the default temperature for NVIDIA Llama 3.3 Nemotron Super 49B v1?
The default temperature for NVIDIA Llama 3.3 Nemotron Super 49B v1 is 0.6, within a valid range of 0 to 1.
What is the default top_p for NVIDIA Llama 3.3 Nemotron Super 49B v1?
The default top_p for NVIDIA Llama 3.3 Nemotron Super 49B v1 is 0.95, with a maximum of 1.
What is the default max_tokens for NVIDIA Llama 3.3 Nemotron Super 49B v1?
The default max_tokens for NVIDIA Llama 3.3 Nemotron Super 49B v1 is 4096, within a valid range of 1 to 16384.

Resources

All NVIDIA models Glossary Full catalog

Llama 3.3 Nemotron Super 49B v1 — JSON

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

/api/v1/models/nvidia/llama-3.3-nemotron-super-49b-v1.json 
{
  "$schema": "https://modelparams.dev/api/v1/schema.json",
  "provider": "nvidia",
  "authType": "api_key",
  "apiSurface": "openai-chat-completions",
  "model": "llama-3.3-nemotron-super-49b-v1",
  "params": [
    {
      "path": "temperature",
      "label": "Temperature",
      "description": "Controls randomness. Lower values make outputs more focused; higher values make them more varied. Not recommended to modify both temperature and top_p in the same call.",
      "group": "sampling",
      "type": "number",
      "default": 0.6,
      "range": {
        "min": 0,
        "max": 1
      }
    },
    {
      "path": "top_p",
      "label": "Top P",
      "description": "Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability. Not recommended to modify both temperature and top_p in the same call.",
      "group": "sampling",
      "type": "number",
      "default": 0.95,
      "range": {
        "max": 1
      }
    },
    {
      "path": "max_tokens",
      "label": "Max tokens",
      "description": "Maximum number of tokens to generate. Generation stops when this limit is reached.",
      "group": "generation_length",
      "type": "integer",
      "default": 4096,
      "range": {
        "min": 1,
        "max": 16384
      }
    },
    {
      "path": "frequency_penalty",
      "label": "Frequency penalty",
      "description": "Penalizes new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.",
      "group": "sampling",
      "type": "number",
      "default": 0,
      "range": {
        "min": -2,
        "max": 2
      }
    },
    {
      "path": "presence_penalty",
      "label": "Presence penalty",
      "description": "Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.",
      "group": "sampling",
      "type": "number",
      "default": 0,
      "range": {
        "min": -2,
        "max": 2
      }
    },
    {
      "path": "seed",
      "label": "Seed",
      "description": "Best-effort deterministic sampling seed. Changing the seed produces a different response with similar characteristics. Fix the seed to reproduce results.",
      "group": "sampling",
      "type": "integer",
      "default": 0,
      "range": {
        "min": 0,
        "max": 18446744073709552000
      }
    },
    {
      "path": "stop",
      "label": "Stop",
      "description": "A string or list of strings where the API will stop generating further tokens. The returned text will not contain the stop sequence.",
      "group": "generation_length",
      "type": "string"
    }
  ]
}

Other NVIDIA models

DeepSeek v4 Flash 0731 7 params View DeepSeek v4 Pro 0813 7 params View Gemma 4 31B IT 7 params View GLiNER PII 4 params View GLM-5.3 Flash 7 params View GPT-OSS 120B 7 params View GPT-OSS 20B 7 params View Kimi K3 7 params View Laguna Xs 2.1 7 params View Llama 3.1 NemoGuard 8B Topic Control 6 params View Llama 3.1 Nemotron Nano 8B v1 7 params View Llama 3.1 Nemotron Safety Guard 8B v3 1 param View Llama 3.1 Nemotron Ultra 253B v1 7 params View Llama 3.3 Nemotron Super 49B v1.5 7 params View MiniMax M3 7 params View Muse Glimmer 30B 7 params View NemoGuard Jailbreak Detect 0 params View Nemotron 3 Nano 30B A3B 5 params View Nemotron 3 Super 120B A12B 7 params View Nemotron 3 Ultra Subscription 6 params View Nemotron 3 Ultra 550B A55B 7 params View Nemotron Content Safety Reasoning 4B 5 params View Nemotron Mini 4B Instruct 6 params View Riva Translate 4B Instruct v1.1 6 params View USDCode Llama 3.1 70B Instruct 4 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/models/openai/gpt-5.5.json
curl https://modelparams.dev/api/v1/models/openai/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