Instructions to use dcostenco/prism-coder-1.7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dcostenco/prism-coder-1.7b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="dcostenco/prism-coder-1.7b",
	filename="prism-aac-1b7-q4km.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use dcostenco/prism-coder-1.7b with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf dcostenco/prism-coder-1.7b:Q8_0
# Run inference directly in the terminal:
llama-cli -hf dcostenco/prism-coder-1.7b:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf dcostenco/prism-coder-1.7b:Q8_0
# Run inference directly in the terminal:
llama-cli -hf dcostenco/prism-coder-1.7b: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 dcostenco/prism-coder-1.7b:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf dcostenco/prism-coder-1.7b: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 dcostenco/prism-coder-1.7b:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dcostenco/prism-coder-1.7b:Q8_0

Use Docker

docker model run hf.co/dcostenco/prism-coder-1.7b:Q8_0

LM Studio
Jan
Ollama
How to use dcostenco/prism-coder-1.7b with Ollama:
```
ollama run hf.co/dcostenco/prism-coder-1.7b:Q8_0
```

Unsloth Studio

How to use dcostenco/prism-coder-1.7b with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for dcostenco/prism-coder-1.7b to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for dcostenco/prism-coder-1.7b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for dcostenco/prism-coder-1.7b to start chatting

How to use dcostenco/prism-coder-1.7b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf dcostenco/prism-coder-1.7b:Q8_0

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "dcostenco/prism-coder-1.7b:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use dcostenco/prism-coder-1.7b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf dcostenco/prism-coder-1.7b: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 dcostenco/prism-coder-1.7b:Q8_0

Run Hermes

hermes

Docker Model Runner
How to use dcostenco/prism-coder-1.7b with Docker Model Runner:
```
docker model run hf.co/dcostenco/prism-coder-1.7b:Q8_0
```

Lemonade

How to use dcostenco/prism-coder-1.7b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dcostenco/prism-coder-1.7b:Q8_0

Run and chat with the model

lemonade run user.prism-coder-1.7b-Q8_0

List all available models

lemonade list

dcostenco commited on 20 days ago

Commit

f5ef1d0

verified ·

1 Parent(s): ce550f2

docs: honest numbers — 14B=98% ties Opus, cascade is fallback not booster

Browse files

Files changed (1) hide show

README.md +17 -16

README.md CHANGED Viewed

@@ -13,31 +13,32 @@ tags:
 - prism-coder
 ---
-# prism-coder:1.7b (v19) — 86% on-device routing
-On-device MCP tool router based on **Qwen/Qwen3-1.7B**. Runs in 1.6 GB RAM at Q4_K_M. Built for iOS / Android / older Macs where larger tiers can't fit.
-## Test results — Prism routing 100-case eval (May 15 2026, 3-seed mean)
 | Category | Score |
 |---|---|
-| **Overall** | **86.3% ± 0.6%** |
 | session_load_context | 100% |
 | session_search_memory | 100% |
-| brave_web_search | 100% |
-| AAC plain-text | **100%** |
-| translate plain-text | 100% |
-| knowledge_search | 43% |
-| session_save_ledger | 71% |
-| session_save_handoff | 87% |
-| **avg latency** | **2.3s** |
-| **invented tools** | 1 |
-**Below the 90% gate** — published for the on-device / cost-sensitive use case, not accuracy-critical work. AAC routing is 100% (life-critical path).
 ## iOS deployment
-GGUF: `prism-aac-1b7-q4km.gguf` (1.0 GB, ~1.6 GB RAM). Integrated via llama.cpp Swift SPM in [Prism AAC](https://github.com/dcostenco/prism-aac).
 ## Usage
@@ -47,9 +48,9 @@ ollama pull dcostenco/prism-coder:1b7
 ## Hardware
-- **iPhone**: A14 Bionic+ (iPhone 12+), ~1.6 GB free RAM
 - **Mac**: any M-series
 ## License
-Apache-2.0 (inherits from Qwen3-1.7B).

 - prism-coder
 ---
+# prism-coder:1.7b (v19) — 88% on-device routing
+On-device MCP tool router. Runs in 1.6 GB RAM at Q4_K_M. Built for iPhone / low-memory devices where the 14B can't load.
+## Routing accuracy — 100-case Prism eval (May 15 2026, 3-seed mean)
 | Category | Score |
 |---|---|
+| **Overall** | **88%** |
+| AAC plain-text | **100%** |
 | session_load_context | 100% |
 | session_search_memory | 100% |
+| knowledge_search | 71% |
+| session_save_ledger | 77% |
+| avg latency | 1.6s |
+| invented tools | 0-2 |
+**Below the 90% gate** — this is the on-device fallback, not the accuracy tier. In production, the [Prism AAC](https://github.com/dcostenco/prism-aac) cascade tries the 14B (98%) first and only falls back to the 1.7B when the 14B can't be loaded.
+AAC routing is 100% — the life-critical path (expressing pain, asking for help) never fails.
+Uses system-prompt engineering only (no LoRA — Q4_K_M quantization erases fine-tuning signal at 1.7B scale).
 ## iOS deployment
+GGUF: `prism-aac-1b7-q4km.gguf` (1.0 GB, ~1.6 GB RAM). Integrated via llama.cpp Swift SPM.
 ## Usage
 ## Hardware
+- **iPhone**: A14+ (iPhone 12+), ~1.6 GB RAM
 - **Mac**: any M-series
 ## License
+Apache-2.0.