Nemotron 4 340B Reward is NVIDIA's language model. 340B reward model that scores response quality for alignment and synthetic data filtering.
Specifications
Canonical IDnvidia-nemotron-4-340b-reward
TypeLanguage
StatusActive
CreatorNVIDIANVIDIA
Input ModalitiesText
Output ModalitiesText
Parameters340B

Capabilities

Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Cheapest Instances to Run It

Cloud GPU instances that can host Nemotron 4 340B Reward, ranked by cheapest on-demand price. The model needs about 816 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.

All clouds
FP16 (full precision)
US Dollar ($)
Instance
Cloud
GPU
VRAM
Price
Cheapest region
Standard_ND96is_MI300X_v5Azure8× AMD Instinct MI300X GPU (192GB)1536 GB$48.00/hr
Standard_ND96isr_MI300X_v5Azure8× AMD Instinct MI300X1536 GB$48.00/hr
p5en.48xlargeAWS8× H2001128 GB$63.30/hr
6 more instances can run Nemotron 4 340B Reward
Unlock the full ranked list and FP8 / INT4 quantization with a CloudPrice subscription.

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Nemotron 4 340B Reward————Current
Nemotron 4 15B————Available
Nemotron 4 340B Instruct————Available
Nemotron 3.5 Super VL1.0M$0.081$0.470Available
Nemotron 3.5 Lightning 30B1.0M——Available
Nemotron Lightning 3.51.0M$0.060$0.160Available
Nemotron 3 Ultra 550B A55B1.0M$0.250$2.15Deprecated
Nemotron 3.5 Content Safety131K$0.200$0.200Available
Nemotron Nano 3 30B A3B Omni Reasoning256K——Available
Nemotron Super 3 120B256K$0.150$0.650Available
Nemotron Nano 3 30B262K$0.060$0.240Available

Model IDs

nvidia_nim/nvidia/nemotron-4-340b-reward
nvidia-nemotron-4-340b-reward