News icon

Batch Inference is now generally available on Runpod Serverless

NVIDIA

NVIDIA Nemotron 3.5 Bolt E4B BF16

NVIDIA's 8B-parameter open language model with a 64K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Language
Text generation
8B · 64K context
See deploy options

NVIDIA Nemotron 3.5 Bolt E4B BF16

Specifications

Provider
NVIDIA
Modality
Language
Task
Text generation
Parameters
8B
Active parameters
Context
64K tokens
Precision
BF16
Weights
16
License
openmdw-1.1
Released
2026-10-04
Endpoint ID
Weights on Hugging Face
nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16

Frequently asked questions

How do I deploy NVIDIA Nemotron 3.5 Bolt E4B BF16 on Runpod?

Deploy the vLLM worker from the Runpod Hub as a Serverless endpoint and set MODEL_NAME to nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16. The endpoint is OpenAI-compatible and scales to zero when idle. You can also run vLLM on a Pod.

What GPU do I need to run NVIDIA Nemotron 3.5 Bolt E4B BF16?

Start with RTX 4090 (24 GB VRAM). The weights take about 16 GB; longer contexts and more concurrent requests need extra memory for the KV cache.

How much does NVIDIA Nemotron 3.5 Bolt E4B BF16 cost on Runpod?

Serverless endpoints and Pods bill per second for the GPU you choose. A Serverless endpoint costs nothing while scaled to zero.

Related models

Z.ai

GLM 5.3

Z.ai's 753B-parameter open language model with a 1M-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Language
753B · 1M context
Text generation
Public Endpoint
Serverless
Qwen

Qwen3.8 2.4T A95B

Qwen's 2400B-parameter mixture-of-experts (95B active) open language model with a 256K-token context, ready to serve across a multi-node Runpod Cluster.

Language
2.4T MoE · 95B active · 256K context
Text generation
Clusters
IBM

IBM Granite 4.0 H Small

IBM Granite-4.0-H-Small is a 32B parameter long-context instruct model.

Language
Text generation
Text generation
Public Endpoint
Moonshot AI

Kimi K3

Moonshot AI's open vision-language model with a 1M-token context, ready to serve across a multi-node Runpod Cluster.

Vision + language
1M context
Text generation
Public Endpoint
Clusters
Microsoft

Mage VL Cowork

Microsoft's 4.7B-parameter open vision-language model with a 256K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Vision + language
4.7B · 256K context
Text generation
Serverless
Baidu

Unlimited OCR 2

Baidu's 3.3B-parameter open vision-language model with a 128K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Vision + language
3.3B · 128K context
Text generation
Serverless
Google

Nano Banana 2

Google's Nano Banana 2 model for AI-powered image editing. Provide up to 14 reference images and a text prompt to produce edited outputs at 1K, 2K, or 4K resolution.

Image
Image editing
Image editing
Public Endpoint
Tongyi Lab

Z Image Turbo

Z-Image is a powerful and highly efficient image generation model family with 6B parameters

Image
Text to image
Text to image
Public Endpoint
Qwen

Qwen Image Edit 2511 LoRA

Qwen series that achieves significant advances in complex text rendering and precise image editing and has LoRA support.

Image
Image editing
Image editing
Public Endpoint
ByteDance

Seedance 2.5 Image-to-Video

ByteDance Seedance 2.5 animates a start image (and optional last frame) into a 4-30 second video up to 4K, driven by a text prompt, with native audio.

Video
Image to video
Image to video
Public Endpoint
ByteDance

Seedance 2.5 Text-to-Video

ByteDance Seedance 2.5 generates 4-30 second videos up to 4K from a text prompt, with native audio and optional reference images, videos and audio.

Video
Text to video
Text to video
Public Endpoint
Pruna

P-Video-Edit

Edit an existing video from a text prompt, optionally guided by up to 4 reference images, by Pruna. Input videos up to 15 seconds.

Video
Image to video
Image to video
Public Endpoint
Resemble AI

Chatterbox Turbo

Resemble AI's Chatterbox Turbo is the fastest open source text to speech model with paralinguistic tags for non-speech sounds with expressive voice cloning capabilities.

Audio
Text to speech
Text to speech
Public Endpoint
MiniMax

Minimax Speech 02 HD

MiniMax is a high-definition text-to-speech model

Audio
Text to speech
Text to speech
Public Endpoint
No items found.
nvidia-nvidia-nemotron-3-5-bolt-e4b-bf16
nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16
Language
2026-10-04 17:00
RTX 4090