Instructions to use rdaum/decisive-dory with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use rdaum/decisive-dory with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf rdaum/decisive-dory:Q8_0
# Run inference directly in the terminal:
llama cli -hf rdaum/decisive-dory:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf rdaum/decisive-dory:Q8_0
# Run inference directly in the terminal:
llama cli -hf rdaum/decisive-dory:Q8_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf rdaum/decisive-dory:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf rdaum/decisive-dory:Q8_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf rdaum/decisive-dory:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf rdaum/decisive-dory:Q8_0

Use Docker

docker model run hf.co/rdaum/decisive-dory:Q8_0

Llama new
LM Studio
Jan

vLLM

How to use rdaum/decisive-dory with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rdaum/decisive-dory"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rdaum/decisive-dory",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rdaum/decisive-dory:Q8_0

Ollama
How to use rdaum/decisive-dory with Ollama:
```
ollama run hf.co/rdaum/decisive-dory:Q8_0
```
Unsloth Desktop

How to use rdaum/decisive-dory with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf rdaum/decisive-dory:Q8_0

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "rdaum/decisive-dory:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use rdaum/decisive-dory with Docker Model Runner:
```
docker model run hf.co/rdaum/decisive-dory:Q8_0
```

Lemonade

How to use rdaum/decisive-dory with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull rdaum/decisive-dory:Q8_0

Run and chat with the model

lemonade run user.decisive-dory-Q8_0

List all available models

lemonade list

Hermes Agent

How to use rdaum/decisive-dory with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf rdaum/decisive-dory:Q8_0

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default rdaum/decisive-dory:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use rdaum/decisive-dory with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf rdaum/decisive-dory:Q8_0

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "rdaum/decisive-dory:Q8_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

rdaum commited on 5 days ago

Commit

9b0be00

verified ·

1 Parent(s): ca51159

Show decision accuracy by task with chart and highlighted table results

Browse files

Files changed (4) hide show

.gitattributes +1 -0
README.md +16 -8
evaluation/decision-comparison.png +3 -0
evaluation/decision-comparison.svg +678 -0

.gitattributes CHANGED Viewed

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 dory-decisions-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
 nvfp4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text

 dory-decisions-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
 nvfp4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
+evaluation/decision-comparison.png filter=lfs diff=lfs merge=lfs -text

README.md CHANGED Viewed

@@ -175,18 +175,26 @@ The BF16 weights have not received the same accuracy evaluation.
 ### Comparison with hosted decision models
 The same 512 questions were submitted to four models through OpenRouter's native decision API on 21 October 2026.
 Each request used the same context, question, answer descriptions, and option order as the local evaluations. Reference answers were not sent.
 Hosted models used their providers' prompt templates and probability settings; no prompts or calibration were fitted for this comparison.
-| Model | Accuracy | Brier ↓ |
-|---|---:|---:|
-| Dory Typed Decisions — Q8_0 / llama.cpp | 77.73% | 0.2599 |
-| Dory Typed Decisions — NVFP4 / Eider | 75.59% | 0.2706 |
-| [Perplexity Decider 1.2](https://openrouter.ai/perplexity/pplx-decider-v1.2) | 74.61% | 0.3178 |
-| [Liquid d1](https://openrouter.ai/liquid/d1) | 66.21% | 0.4156 |
-| [TypeSafe Jev 1.14](https://openrouter.ai/typesafe/jev-1.14) | 66.02% | 0.3843 |
-| [Kev 4B](https://openrouter.ai/jaredpalmer/kev-4b) | 42.58% | 0.5134 |
 This test measures new examples from task families included in this fine-tune's training. It does not establish a general ranking of decision models.
 The hosted models' training overlap with these sources is unknown.

 ### Comparison with hosted decision models
+![Accuracy by task for base Dory, fine-tuned Dory, Jev, Kev, Liquid d1, and Perplexity Decider.](evaluation/decision-comparison.png)
 The same 512 questions were submitted to four models through OpenRouter's native decision API on 21 October 2026.
 Each request used the same context, question, answer descriptions, and option order as the local evaluations. Reference answers were not sent.
 Hosted models used their providers' prompt templates and probability settings; no prompts or calibration were fitted for this comparison.
+| Model | State tracking | Planning | Human judgements | Combined | Brier ↓ |
+|---|---:|---:|---:|---:|---:|
+| Base Dory — Q8_0 / llama.cpp | 31.77% | 38.02% | 49.22% | 38.48% | 0.6736 |
+| Base Dory — NVFP4 / Eider | 29.69% | 42.71% | 47.66% | 39.06% | 0.6865 |
+| Dory Typed Decisions — Q8_0 / llama.cpp | **76.04%** | **96.88%** | 51.56% | **77.73%** | **0.2599** |
+| Dory Typed Decisions — NVFP4 / Eider | 75.52% | 92.19% | 50.78% | 75.59% | 0.2706 |
+| [TypeSafe Jev 1.14](https://openrouter.ai/typesafe/jev-1.14) | 64.58% | 74.48% | 55.47% | 66.02% | 0.3843 |
+| [Kev 4B](https://openrouter.ai/jaredpalmer/kev-4b) | 38.54% | 36.46% | 57.81% | 42.58% | 0.5134 |
+| [Liquid d1](https://openrouter.ai/liquid/d1) | 58.33% | 79.17% | **58.59%** | 66.21% | 0.4156 |
+| [Perplexity Decider 1.2](https://openrouter.ai/perplexity/pplx-decider-v1.2) | 72.92% | 88.02% | 57.03% | 74.61% | 0.3178 |
+Accuracy columns show percentages; **bold** marks the best result in each column. Combined accuracy weights all 512 questions equally.
+All four hosted models score higher than this fine-tune on human judgements. Jev scores 55.47%, compared with 51.56% for Q8_0 and 50.78% for NVFP4.
+The fine-tune scores higher on state tracking and planning.
 This test measures new examples from task families included in this fine-tune's training. It does not establish a general ranking of decision models.
 The hosted models' training overlap with these sources is unknown.

evaluation/decision-comparison.png ADDED Viewed

Git LFS Details

SHA256: 3ee042a648080caf7b1fe935427ff6d409bb7b76280d492b7ca66e8629d95f61
Pointer size: 131 Bytes
Size of remote file: 229 kB

evaluation/decision-comparison.svg ADDED Viewed