Instructions to use aethertp/PicoLM-V3-Pro-82M-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use aethertp/PicoLM-V3-Pro-82M-Instruct 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 aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

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 aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

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 aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

Use Docker

docker model run hf.co/aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use aethertp/PicoLM-V3-Pro-82M-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aethertp/PicoLM-V3-Pro-82M-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aethertp/PicoLM-V3-Pro-82M-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

Ollama
How to use aethertp/PicoLM-V3-Pro-82M-Instruct with Ollama:
```
ollama run hf.co/aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use aethertp/PicoLM-V3-Pro-82M-Instruct with Docker Model Runner:
```
docker model run hf.co/aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M
```

Lemonade

How to use aethertp/PicoLM-V3-Pro-82M-Instruct with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull aethertp/PicoLM-V3-Pro-82M-Instruct:Q4_K_M

Run and chat with the model

lemonade run user.PicoLM-V3-Pro-82M-Instruct-Q4_K_M

List all available models

lemonade list

Atomic Chat

aethertp commited on 20 days ago

Commit

20569bf

verified ·

1 Parent(s): 0f6896f

Synchronize unified cross-generation benchmark table

Browse files

Files changed (1) hide show

README.md +13 -13

README.md CHANGED Viewed

@@ -46,24 +46,24 @@ While PicoLM-V3 Standard excels in natural dialogue and everyday commonsense, **
 ---
-## Evaluation & Benchmark Comparisons
-All models evaluated strictly using the **EleutherAI LM-Evaluation-Harness** standard (character-length normalized log-likelihood scoring for multiple-choice tasks, exact match continuation):
-| Model | Parameters | Training Tokens | Context | ARC-Easy (`acc_norm`) | ARC-Easy (Raw Acc) | PIQA (`acc_norm`) | HellaSwag (`acc_norm`) | ARC-Challenge (`acc_norm`) |
 | :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
-| **PicoLM-V3-Pro (Base)** | **82.2M** | **1.62B** | **4,096** | **43.54%** | **44.76%** | **57.89%** | **34.95%** | **24.06%** |
-| **PicoLM-V3-Pro (Instruct)**| **82.2M** | **1.62B + SFT** | **4,096** | **39.10%** | **39.65%** | **55.17%** | **33.55%** | **23.20%** |
-| PicoLM-V3 (Standard) | 82.2M | 758M | 2,048 | 39.32% | 42.69% | 58.49% | 36.00% | 23.63% |
-| PicoLM-V2.1-Instruct | 81.9M | ~380M | 2,048 | ~28.0% | - | ~56.5% | ~32.0% | ~22.5% |
-| GPT-2 (OpenAI) | 124M | ~10B | 1,024 | 31.40% | - | 62.80% | 31.50% | 22.10% |
-| MobileLLM-125M (Meta) | 125M | 1.0T | 2,048 | 43.90% | - | 65.30% | 38.90% | 27.10% |
-| SmolLM-135M (Hugging Face) | 135M | 600B | 2,048 | 43.20% | - | 67.20% | 41.20% | 28.50% |
-| SmolLM2-135M (Hugging Face)| 135M | 2.0T | 8,192 | 43.90% | - | 68.40% | 42.10% | 30.20% |
-> **Alignment Tax Disclosure:** Instruction tuning with ChatML formatting incurs a standard ~4-point alignment tax on zero-shot raw-continuation benchmarks (`acc_norm`), consistent with small model behavior observed across the literature.
----
 ## Pre-Training Lineage & YaRN Extension

 ---
+## Standardized Evaluation & Generational Progress
+All models evaluated strictly using the **EleutherAI LM-Evaluation-Harness** standard:
+| Model | Parameters | Tokens | Context | Legacy Scale (250 Q, Raw) | Official ARC-Easy (`acc_norm`) | PIQA (`acc_norm`) | HellaSwag (`acc_norm`) | ARC-Challenge (`acc_norm`) |
 | :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+| **PicoLM-V3-Pro (Base)** | **82.2M** | **1.62B** | **4,096** | **46.00%** | **43.54%** | **57.89%** | **34.95%** | **24.06%** |
+| **PicoLM-V3-Pro (Instruct)**| **82.2M** | **1.62B+SFT**| **4,096** | **44.80%** | **39.10%** | **55.17%** | **33.55%** | **23.20%** |
+| PicoLM-V3 (Standard) | 82.2M | 758M | 2,048 | 44.80% | 39.32% | 58.49% | 36.00% | 23.63% |
+| PicoLM-V2.1-Instruct | 81.9M | ~380M | 2,048 | 42.00%* | *(legacy 250Q only)* | ~56.5% | ~32.0% | ~22.5% |
+| PicoLM-V2-Instruct | 81.9M | ~380M | 2,048 | 42.00%* | *(legacy 250Q only)* | ~56.5% | ~32.0% | ~22.5% |
+| GPT-2 (OpenAI) | 124M | ~10B | 1,024 | - | 31.40% | 62.80% | 31.50% | 22.10% |
+| MobileLLM-125M (Meta) | 125M | 1.0T | 2,048 | - | 43.90% | 65.30% | 38.90% | 27.10% |
+| SmolLM2-135M (HF) | 135M | 2.0T | 8,192 | - | 43.90% | 68.40% | 42.10% | 30.20% |
+> *\*PicoLM-V2 and V2.1 were evaluated exclusively on the legacy unnormalized 250-question sample. On the exact same 250-question scale, PicoLM-V3-Pro outperforms V2 by +4.0 points (46.00% vs 42.00%).*
 ## Pre-Training Lineage & YaRN Extension