Jump to Content

API Reference

Introduction

Models

Large Language models

LLM APIs
01-ai / yi-large
- Creates a model response for the given chat conversation.post
abacusai / dracarys-llama-3.1-70b-instruct
- Creates a model response for the given chat conversation.post
ai21labs / jamba-1.5-large-instruct
- Creates a model response for the given chat conversation.post
ai21labs / jamba-1.5-mini-instruct
- Creates a model response for the given chat conversation.post
aisingapore / sea-lion-7b-instruct
- Create a chat completionpost
baichuan-inc / baichuan2-13b-chat
- Creates a model response for the given chat conversation.post
bigcode / starcoder2-7b
- Create Completionpost
bigcode / starcoder2-15b
- Create Completionpost
databricks / dbrx-instruct
- Create a chat completionpost
deepseek-ai / deepseek-r1
- Creates a model response for the given chat conversation.post
deepseek-ai / deepseek-r1-distill-llama-8b
- Creates a model response for the given chat conversation.post
deepseek-ai / deepseek-r1-distill-qwen-7b
- Creates a model response for the given chat conversation.post
deepseek-ai / deepseek-r1-distill-qwen-14b
- Creates a model response for the given chat conversation.post
deepseek-ai / deepseek-r1-distill-qwen-32b
- Creates a model response for the given chat conversation.post
google / codegemma-1.1-7b
- Creates a model response for the given chat conversation.post
google / codegemma-7b
- Create a chat completionpost
google / gemma-2b
- Create a chat completionpost
google / gemma-7b
- Create a chat completionpost
google / gemma-2-2b-it
- Creates a model response for the given chat conversation.post
google / gemma-2-9b-it
- Creates a model response for the given chat conversation.post
google / gemma-2-27b-it
- Creates a model response for the given chat conversation.post
google / gemma-3-1b-it
- Creates a model response for the given chat conversation.post
google / recurrentgemma-2b
- Create a chat completionpost
google / shieldgemma-9b
- Creates a model response for the given chat conversation.post
gotocompany / gemma-2-9b-cpt-sahabatai-instruct
- Creates a model response for the given chat conversation.post
ibm / granite-3.0-3b-a800m-instruct
- Creates a model response for the given chat conversation.post
ibm / granite-3.0-8b-instruct
- Creates a model response for the given chat conversation.post
ibm / granite-3_3-8b-instruct
- Creates a model response for the given chat conversation.post
ibm / granite-34b-code-instruct
- Creates a model response for the given chat conversation.post
ibm / granite-8b-code-instruct
- Creates a model response for the given chat conversation.post
ibm / granite-guardian-3.0-8b
- Creates a model response for the given chat conversation.post
igenius / colosseum_355b_instruct_16k
- Creates a model response for the given chat conversation.post
igenius / italia_10b_instruct_16k
- Creates a model response for the given chat conversation.post
institute-of-science-tokyo / llama-3.1-swallow-70b-instruct-v01
- Creates a model response for the given chat conversation.post
institute-of-science-tokyo / llama-3.1-swallow-8b-instruct-v0.1
- Creates a model response for the given chat conversation.post
marin / marin-8b-instruct
- Creates a model response for the given chat conversation.post
mediatek / breeze-7b-instruct
- Creates a model response for the given chat conversation.post
meta / codellama-70b
- Create a chat completionpost
meta / llama2-70b
- Create a chat completionpost
meta / llama3-8b
- Creates a chat completionpost
meta / llama3-70b
- Creates a chat completionpost
meta / llama-3.1-8b-instruct
- Creates a model response for the given chat conversation.post
meta / llama-3.1-70b-instruct
- Creates a model response for the given chat conversation.post
meta / llama-3.1-405b-instruct
- Creates a model response for the given chat conversation.post
meta / llama-3.2-1b-instruct
- Creates a model response for the given chat conversation.post
meta / llama-3.2-3b-instruct
- Creates a model response for the given chat conversation.post
meta / llama-3.3-70b-instruct
- Creates a model response for the given chat conversation.post
microsoft / phi-3-medium-128k-instruct
- Creates a model response for the given chat conversation.post
microsoft / phi-3-medium-4k-instruct
- Creates a chat completionpost
microsoft / phi-3-mini-128k-instruct
- Creates a model response for the given chat conversation.post
microsoft / phi-3-mini-4k-instruct
- Creates a model response for the given chat conversation.post
microsoft / phi-3-small-128k-instruct
- Creates a chat completionpost
microsoft / phi-3-small-8k-instruct
- Create a chat completionpost
microsoft / phi-3.5-mini
- Creates a model response for the given chat conversation.post
microsoft / phi-3.5-moe-instruct
- Creates a model response for the given chat conversation.post
microsoft / phi-4-mini-instruct
- Creates a model response for the given chat conversation.post
mistralai / codestral-22b-instruct-v0.1
- Creates a model response for the given chat conversation.post
mistralai / mamba-codestral-7b-v0.1
- Creates a model response for the given chat conversation.post
mistralai / mistral-2-large-instruct
- Creates a model response for the given chat conversation.post
mistralai / mathstral-7b-v01
- Creates a model response for the given chat conversation.post
mistralai / mistral-7b-instruct
- Create a chat completionpost
mistralai / mistral-7b-instruct-v0.3
- Creates a model response for the given chat conversation.post
mistralai / mixtral-8x7b-instruct
- Create a chat completionpost
mistralai / mixtral-8x22b-instruct
- Create a chat completionpost
mistralai / mistral-large
- Create a chat completionpost
mistralai / mistral-small-24b-instruct
- Creates a model response for the given chat conversation.post
nvidia / llama3-chatqa-1.5-8b
- Creates a model response for the given chat conversation.post
nvidia / llama-3.1-nemoguard-8b-content-safety
- Creates a model response for the given chat conversation.post
nvidia / llama-3.1-nemoguard-8b-topic-control
- Creates a model response for the given chat conversation.post
nvidia/llama-3.1-nemotron-nano-8b-v1
- Creates a model response for the given chat conversation.post
nvidia / llama-3.1-nemotron-51b-instruct
- Creates a model response for the given chat conversation.post
nvidia/llama-3.1-nemotron-70b-instruct
- Creates a model response for the given chat conversation.post
nvidia / llama-3.1-nemotron-70b-reward
- Creates a model response for the given chat conversation.post
nvidia / llama-3_1-nemotron-ultra-253b-v1
- Creates a model response for the given chat conversation.post
nvidia/llama-3.3-nemotron-super-49b-v1
- Creates a model response for the given chat conversation.post
nvidia / llama3-chatqa-1.5-70b
- Creates a model response for the given chat conversation.post
nvidia / mistral-nemo-minitron-8b-base
- Create Completionpost
nvidia / mistral-nemo-minitron-8b-8k-instruct
- Creates a model response for the given chat conversation.post
nvidia / nemoguard-jailbreak-detect
- Classify text for jailbreak attempt.post
nvidia / nemotron-4-340b-instruct
- Creates a model response for the given chat conversation.post
nvidia / nemotron-4-340b-reward
- Creates a model response for the given chat conversation.post
nvidia / nemotron-4-mini-hindi-4b-instruct
- Creates a model response for the given chat conversation.post
nvidia / nemotron-mini-4b-instruct
- Creates a model response for the given chat conversation.post
nvidia / usdcode
- Creates a model response for the given chat conversation.post
nvidia / usdsearch
- Search Postpost
nv-mistralai / mistral-nemo-12b-instruct
- Creates a model response for the given chat conversation.post
qwen / qwen2-7b-instruct
- Creates a model response for the given chat conversation.post
qwen / qwen2.5-7b-instruct
- Creates a model response for the given chat conversation.post
qwen / qwen2.5-coder-7b-instruct
- Creates a model response for the given chat conversation.post
qwen / qwen2.5-coder-32b-instruct
- Creates a model response for the given chat conversation.post
qwen / qwen3-235b-a22b
- Creates a model response for the given chat conversation.post
qwen / qwq-32b
- Creates a model response for the given chat conversation.post
rakuten / rakutenai-7b-chat
- Creates a model response for the given chat conversation.post
rakuten / rakutenai-7b-instruct
- Creates a model response for the given chat conversation.post
seallms / seallm-7b-v2.5
- Create a chat completionpost
snowflake / arctic
- Create a chat completionpost
tokyotech-llm / llama-3-swallow-70b-instruct-v01
- Creates a model response for the given chat conversation.post
thudm / chatglm3-6b
- Creates a model response for the given chat conversation.post
tiiuae / falcon3-7b-instruct
- Creates a model response for the given chat conversation.post
upstage / solar-10.7b-instruct
- Creates a model response for the given chat conversation.post
utter-project / eurollm-9b-instruct
- Creates a model response for the given chat conversation.post
writer / palmyra-creative-122b
- Creates a model response for the given chat conversation.post
writer / palmyra-fin-70b-32k
- Creates a model response for the given chat conversation.post
writer / palmyra-med-70b-32k
- Create a chat completionpost
writer / palmyra-med-70b
- Create a chat completionpost
yentinglin / llama-3-taiwan-70b-instruct
- Creates a model response for the given chat conversation.post
zyphra/zamba2-7b-instruct
- Creates a model response for the given chat conversation.post

Retrieval

Retrieval APIs
baai / bge-m3
- Creates an embedding vector from the input text.post
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / embed-qa-4
- Create embedding vectorpost
nvidia / llama-3.2-nv-embedqa-1b-v1
- Creates an embedding vector from the input text.post
nvidia / llama-3.2-nv-embedqa-1b-v2
- Creates an embedding vector from the input text.post
nvidia / llama-3.2-nv-rerankqa-1b-v1
- Rank passages by their relation to a query.post
nvidia / llama-3.2-nv-rerankqa-1b-v2
- Rank passages by their relation to a query.post
nvidia / nvclip
- Creates an embedding vector representing the input text or image.post
nvidia / nv-embed-v1
- Creates an embedding vector from the input text.post
nvidia / nv-embedcode-7b-v1
- Creates an embedding vector from the input text.post
nvidia / nv-embedqa-e5-v5
- Creates an embedding vector from the input text.post
nvidia / nv-embedqa-mistral-7b-v2
- Creates an embedding vector from the input text.post
nvidia / nv-rerankqa-mistral-4b-v3
- Rank passages by their relation to a query.post
nvidia / rerank-qa-mistral-4b
- Create rankingpost
snowflake / arctic-embed-l
- Creates an embedding vector from the input text.post
- Gets the result of an earlier function invocation request that returned a status of 202.get

Visual Models

Visual Models APIs
black-forest-labs / flux.1-dev
- Inferpost
black-forest-labs / flux_1-schnell
- Inferpost
briaai / bria-2.3
- Request generationpost
google / gemma-3-27b-it
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
hive / ai-generated-image-detection
- Request response from the modelpost
hive / deepfake-image-detection
- Inferpost
meta / sam2
- Run inference on the input image/video for a given prompt.post
- Gets the result of an earlier function invocation request that returned a status of 202.get
microsoft / phi-4-multimodal-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
mistralai / mistral-medium-3-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
mistralai / mistral-small-3_1-24b-instruct-2503
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / bevformer
- Post V1 Bevformer Processpost
nvidia / consistory
- Request generationpost
nvidia / cosmos-1.0-7b-diffusion-text2world
nvidia / nemoretriever-parse
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / nv-dinov2
- Run inference on the input image.post
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / nv-grounding-dino
- Run inference on the input image/video for a given text promptpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / ocdrnet
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / retail-object-detection
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / sparsedrive
- Post V1 Sparsedrive Inferencepost
nvidia / vila
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / visual-changenet
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
stabilityai / stable-diffusion-3-medium
- Request generationpost
stabilityai / sdxl-turbo
- Request generationpost
stabilityai / stable-diffusion-xl
- Request generationpost
stabilityai / stable-video-diffusion
- Request generationpost

multimodAl

Multimodal APIs
adept / fuyu-8b
- Request response from the modelpost
- Status pollingget
google / deplot
- Request response from the modelpost
- Status pollingget
google / paligemma
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
meta/llama-3.2-11b-vision-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
meta/llama-3.2-90b-vision-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
meta / llama-4-maverick-17b-128e-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
meta / llama-4-scout-17b-16e-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
microsoft / florence-2
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
microsoft / kosmos-2
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
microsoft / phi-3-vision-128k-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
microsoft / phi-3.5-vision-instruct
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get
nvidia / neva-22b
- Request response from the modelpost
- Gets the result of an earlier function invocation request that returned a status of 202.get

Healthcare

Healthcare APIs
arc / evo2-40b
- Generate DNA sequencespost
colabfold / msa-search
- Nim Api Post Call Msa Search Postpost
deepmind / alphafold2
- Predict Structure From Sequence Postpost
deepmind / alphafold2-multimer
- Predict Structure From Sequence Postpost
ipd / proteinmpnn
- Predict amino acid sequencespost
ipd / rfdiffusion
- Run RFdiffusion Protein Generationpost
meta / esmfold
- Predict protein structure (alignment-free)post
meta / esm2-650m
- Protein Embeddingspost
mit / diffdock
- Molecular Docking Pose Generationpost
nvidia / deepvariant
- Run Parabricks Universal Variant Callingpost
nvidia / fq2bam
- Run Parabricks fq2bam to align sequence readspost
nvidia / genmol
- Molecular Generationpost
nvidia / maisi
- Generate Imagepost
nvidia / molmim
- Perform molecule generationpost
nvidia / vista3d
- Run Inferencepost
openfold / openfold2
- Nim Api Post Call Monomer Structure From Msa And Templatepost

route optimization

Route Optimization APIs
nvidia / cuOpt
- Submit to solverpost
- Status pollingget

climate simulation

Climate Simulation APIs
nvidia / corrdiff
- Inferencepost
nvidia / fourcastnet
- Runs FourCastNet inference.post

Route Optimization APIs

Overview

Models

nvidia

Model	Endpoint
nvidia / cuOpt	Submit to solver (cuOpt)
nvidia / cuOpt	Status polling (cuOpt)

Table of Contents
- Overview
  - Models

Terms of Use | Privacy Policy | Manage My Privacy | Accessibility | Corporate Policies | Product Security | Contact

Copyright© 2024 NVIDIA Corporation