# Ripple 65B Preview

> Ripple 65B Preview is Ricursive Intelligence's first open-weight model, a 65B-parameter dense transformer with native tool calling and reasoning. It has a 1,048,576-token context window and was post-trained mainly for agentic coding. Weights are Apache 2.0.

**ID**: `ricursive/ripple-65b-preview`  
**Creator**: ricursive  
**Category**: text  
**Context**: 1M tokens  
**Pricing**: $0.10 / $0.30 per 1M · billed by your provider (BYOK)  
**Released**: 2026-11-04  
**Links**: [Weights on Hugging Face](https://huggingface.co/ricursiveAI/Ripple-65B-Preview) · [Announcement](https://www.ricursive.com/ripple) · [AnyRouter release](https://anyrouter.dev/blog/releases/2026-11-15-ripple-65b-preview)  
**Web page**: https://anyrouter.dev/model/ricursive/ripple-65b-preview

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

**Tokenizer**: Other

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

**Aliases**: `ricursiveai/ripple-65b-preview`

## 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": "ricursive/ripple-65b-preview",
  "messages": [
    {
      "role": "user",
      "content": "Say hi in 3 words."
    }
  ]
}'
```

## Providers

| Provider | Input | Output |
| --- | --- | --- |
| openrouter-byok | $0.10/M (list) | $0.30/M (list) |
| nousresearch-byok | $0.10/M (list) | $0.30/M (list) |
| gmicloud-byok | $0.20/M (list) | $0.60/M (list) |
| nano-gpt-byok | $0.10/M (list) | $0.30/M (list) |
| edenai-byok | $0.10/M (list) | $0.30/M (list) |

BYOK routes are billed by your own provider key — AnyRouter charges $0. Rates marked (list) are the provider's published list price.

## Model card

- Architecture: dense [2]
- Total parameters: 65B [2]
- Active parameters: 65B [2]
- Native context: 1048576 tokens [2]
- License: Apache-2.0, open weights (https://huggingface.co/ricursiveAI/Ripple-65B-Preview) [2]

### Strengths [1,2]

- Reports 80.52 on Terminal-Bench 2.1, which Ricursive presents as the top score among US open-weight models (vendor-reported).
- Dense 65B model that fits on a single DGX Spark class machine, according to the launch post.
- Native tool calling with up to 1M tokens of context.

### Known limits [2]

- The Preview had relatively light post-training with a focus on agentic coding.
- All benchmark figures are vendor-reported, and some come from in-house evaluations.

### Use cases [2]

- Agentic coding and terminal tasks.
- Long-context work with tool calling.

### API notes [2]

- Native tool calling and reasoning, as stated on the model card.

### Links

- announcement: https://www.ricursive.com/ripple
- model card: https://huggingface.co/ricursiveAI/Ripple-65B-Preview

### Sources

1. [Ripple-65B Preview, a Frontier Open-Weight Model](https://www.ricursive.com/ripple) (accessed 2026-11-15)
2. [ricursiveAI/Ripple-65B-Preview model card](https://huggingface.co/ricursiveAI/Ripple-65B-Preview) (accessed 2026-11-15)

## Source

- Upstream docs: https://www.ricursive.com/ripple