# Nemotron 3.5 Bolt E4B

> NVIDIA Nemotron 3.5 Bolt E4B is a small language model post-trained from Gemma 4 E4B. It improves multi-turn tool use, instruction following and input grounding for fast voice and device assistants, interactive personas and embedded interfaces, with low memory use and low time to first token.

**ID**: `nvidia/nemotron-3.5-bolt-e4b`  
**Creator**: nvidia  
**Category**: text  
**Context**: 66K tokens  
**Input**: $0  
**Output**: $0  
**Released**: 2026-10-23  
**Web page**: https://anyrouter.dev/model/nvidia/nemotron-3.5-bolt-e4b

**Input modalities**: text  
**Output modalities**: text  

**Capabilities**: chat, streaming, function-calling, reasoning

## Usage

**Endpoint**: `POST https://anyrouter.dev/api/v1/chat/completions`

```bash
curl https://anyrouter.dev/api/v1/chat/completions \
  -H "Authorization: Bearer $ANYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "nvidia/nemotron-3.5-bolt-e4b",
  "messages": [
    {
      "role": "user",
      "content": "Say hi in 3 words."
    }
  ]
}'
```

## Providers

| Provider | Input | Output |
| --- | --- | --- |
| nvidia | $0 | $0 |
| nvidia-byok | $0 | $0 |

## Model card

- Architecture: dense [1]
- Total parameters: 8B [1]
- Native context: 65536 tokens [1]
- Training cutoff: 2026-05 [1]
- License: OpenMDW-1.1, open weights (https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16) [1]

### Strengths [1]

- Built for low memory use and low time to first token (vendor-stated).
- Improves multi-turn tool use, instruction following and input grounding over Gemma 4 E4B (vendor-reported).

### Known limits [1]

- Context is 64K tokens for input and 64K for output.
- Text in, text out only.

### Use cases [1]

- Voice and device assistants
- Interactive personas
- Embedded and edge interfaces

### API notes [1]

- Thinking is controlled with the chat-template flag `enable_thinking`.
- Recommended sampling is temperature 1.0 and top_p 0.95 in both thinking modes.

### Links

- announcement: https://research.nvidia.com/labs/nemotron/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB/
- model card: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB
- paper: https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB-Technical-Report.pdf
- docs: https://build.nvidia.com/nvidia/nemotron-3.5-bolt-e4b

### Sources

1. [NVIDIA-Nemotron-3.5-Bolt-E4B-4GB model card](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB) (accessed 2026-10-24)
2. [NVIDIA Nemotron 3.5 Bolt E4B 4GB](https://research.nvidia.com/labs/nemotron/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB/) (accessed 2026-10-24)

## Source

- Upstream docs: https://build.nvidia.com/nvidia/nemotron-3.5-bolt-e4b