bonigopalan/ grove-explore-base

3 Downloads Updated 13 hours ago

The local code-locator delegate for grove’s delegated exploration mode.

ollama run bonigopalan/grove-explore-base

curl http://localhost:11434/api/chat \
  -d '{
    "model": "bonigopalan/grove-explore-base",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='bonigopalan/grove-explore-base',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'bonigopalan/grove-explore-base',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Models

Name

3 models

Size / Usage

Context

Input

grove-explore-base:latest

2.8GB · 256K context window · Text · 13 hours ago

grove-explore-base:latest

2.8GB

256K

Text

grove-explore-base:Q4_K_M

2.8GB · 256K context window · Text · 13 hours ago

grove-explore-base:Q4_K_M

2.8GB

256K

Text

grove-explore-base:q8_0

4.6GB · 256K context window · Text · 13 hours ago

grove-explore-base:q8_0

4.6GB

256K

Text

Readme

grove-explore-base

The local code-locator delegate for grove’s delegated exploration mode (grove init --as mcp-llm). grove’s inner Rust explorer drives a short tool-calling loop against this model over an OpenAI-compatible endpoint and returns validated file:line citations.

Untrained base. Off-the-shelf Qwen3.5-4B (Alibaba/Qwen), self-converted to GGUF and quantized — not a grove fine-tune.

Pull

ollama pull bonigopalan/grove-explore-base:q4_k_m   # default, 2.78 GB
ollama pull bonigopalan/grove-explore-base:q8_0     # baseline, 4.6 GB

Point grove at it — `.grove/explore.json`

{
  "provider": "ollama",
  "base_url": "http://localhost:11434/v1",
  "model": "bonigopalan/grove-explore-base:q4_k_m",
  "steering": "strict"
}

Serving notes

Thinking is ON (default on ollama ≥ 0.32) — required; thinking off yields empty tool-calls.
Baked-in params: temperature 0, num_ctx 24576.

Quant	Size	Eval (sheet-coverage, n=347)	Role
`q4_k_m`	2.78 GB	80.6	default, memory-lean
`q8_0`	4.6 GB	82.1	canonical baseline

Requires ollama ≥ 0.32 (earlier engines reject this qwen35 GGUF). Apache-2.0, inherited from upstream Qwen3.5-4B.

# grove-explore-base

The local **code-locator delegate** for [grove](https://github.com/Entelligentsia/grove)'s
delegated exploration mode (`grove init --as mcp-llm`). grove's inner Rust
explorer drives a short tool-calling loop against this model over an
OpenAI-compatible endpoint and returns validated `file:line` citations.

> **Untrained base.** Off-the-shelf **Qwen3.5-4B** (Alibaba/Qwen), self-converted
> to GGUF and quantized — not a grove fine-tune.

## Pull

ollama pull bonigopalan/grove-explore-base:q4_k_m   # default, 2.78 GB
    ollama pull bonigopalan/grove-explore-base:q8_0     # baseline, 4.6 GB

## Point grove at it — `.grove/explore.json`

{
      "provider": "ollama",
      "base_url": "http://localhost:11434/v1",
      "model": "bonigopalan/grove-explore-base:q4_k_m",
      "steering": "strict"
    }

## Serving notes

- **Thinking is ON** (default on ollama ≥ 0.32) — required; thinking off yields
  empty tool-calls.
- Baked-in params: `temperature 0`, `num_ctx 24576`.

| Quant | Size | Eval (sheet-coverage, n=347) | Role |
|---|---|---|---|
| `q4_k_m` | 2.78 GB | 80.6 | default, memory-lean |
| `q8_0` | 4.6 GB | 82.1 | canonical baseline |

Requires **ollama ≥ 0.32** (earlier engines reject this qwen35 GGUF).
Apache-2.0, inherited from upstream Qwen3.5-4B.

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)