Osmosis-Structure-0.6B GGUF Models

Model Generation Details

This model was generated using llama.cpp at commit f5cd27b7.

Choosing the Right Model Format

Selecting the correct model format depends on your hardware capabilities and memory constraints.

BF16 (Brain Float 16) – Use if BF16 acceleration is available

A 16-bit floating-point format designed for faster computation while retaining good precision.
Provides similar dynamic range as FP32 but with lower memory usage.
Recommended if your hardware supports BF16 acceleration (check your device's specs).
Ideal for high-performance inference with reduced memory footprint compared to FP32.

📌 Use BF16 if:
✔ Your hardware has native BF16 support (e.g., newer GPUs, TPUs).
✔ You want higher precision while saving memory.
✔ You plan to requantize the model into another format.

📌 Avoid BF16 if:
❌ Your hardware does not support BF16 (it may fall back to FP32 and run slower).
❌ You need compatibility with older devices that lack BF16 optimization.

F16 (Float 16) – More widely supported than BF16

A 16-bit floating-point high precision but with less of range of values than BF16.
Works on most devices with FP16 acceleration support (including many GPUs and some CPUs).
Slightly lower numerical precision than BF16 but generally sufficient for inference.

📌 Use F16 if:
✔ Your hardware supports FP16 but not BF16.
✔ You need a balance between speed, memory usage, and accuracy.
✔ You are running on a GPU or another device optimized for FP16 computations.

📌 Avoid F16 if:
❌ Your device lacks native FP16 support (it may run slower than expected).
❌ You have memory limitations.

Quantized Models (Q4_K, Q6_K, Q8, etc.) – For CPU & Low-VRAM Inference

Quantization reduces model size and memory usage while maintaining as much accuracy as possible.

Lower-bit models (Q4_K) → Best for minimal memory usage, may have lower precision.
Higher-bit models (Q6_K, Q8_0) → Better accuracy, requires more memory.

📌 Use Quantized Models if:
✔ You are running inference on a CPU and need an optimized model.
✔ Your device has low VRAM and cannot load full-precision models.
✔ You want to reduce memory footprint while keeping reasonable accuracy.

📌 Avoid Quantized Models if:
❌ You need maximum accuracy (full-precision models are better for this).
❌ Your hardware has enough VRAM for higher-precision formats (BF16/F16).

Very Low-Bit Quantization (IQ3_XS, IQ3_S, IQ3_M, Q4_K, Q4_0)

These models are optimized for extreme memory efficiency, making them ideal for low-power devices or large-scale deployments where memory is a critical constraint.

IQ3_XS: Ultra-low-bit quantization (3-bit) with extreme memory efficiency.
- Use case: Best for ultra-low-memory devices where even Q4_K is too large.
- Trade-off: Lower accuracy compared to higher-bit quantizations.
IQ3_S: Small block size for maximum memory efficiency.
- Use case: Best for low-memory devices where IQ3_XS is too aggressive.
IQ3_M: Medium block size for better accuracy than IQ3_S.
- Use case: Suitable for low-memory devices where IQ3_S is too limiting.
Q4_K: 4-bit quantization with block-wise optimization for better accuracy.
- Use case: Best for low-memory devices where Q6_K is too large.
Q4_0: Pure 4-bit quantization, optimized for ARM devices.
- Use case: Best for ARM-based devices or low-memory environments.

Summary Table: Model Format Selection

Model Format	Precision	Memory Usage	Device Requirements	Best Use Case
BF16	Highest	High	BF16-supported GPU/CPUs	High-speed inference with reduced memory
F16	High	High	FP16-supported devices	GPU inference when BF16 isn't available
Q4_K	Medium Low	Low	CPU or Low-VRAM devices	Best for memory-constrained environments
Q6_K	Medium	Moderate	CPU with more memory	Better accuracy while still being quantized
Q8_0	High	Moderate	CPU or GPU with enough VRAM	Best accuracy among quantized models
IQ3_XS	Very Low	Very Low	Ultra-low-memory devices	Extreme memory efficiency and low accuracy
Q4_0	Low	Low	ARM or low-memory devices	llama.cpp can optimize for ARM devices

Included Files & Details

`Osmosis-Structure-0.6B-bf16.gguf`

Model weights preserved in BF16.
Use this if you want to requantize the model into a different format.
Best if your device supports BF16 acceleration.

`Osmosis-Structure-0.6B-f16.gguf`

Model weights stored in F16.
Use if your device supports FP16, especially if BF16 is not available.

`Osmosis-Structure-0.6B-bf16-q8_0.gguf`

Output & embeddings remain in BF16.
All other layers quantized to Q8_0.
Use if your device supports BF16 and you want a quantized version.

`Osmosis-Structure-0.6B-f16-q8_0.gguf`

Output & embeddings remain in F16.
All other layers quantized to Q8_0.

`Osmosis-Structure-0.6B-q4_k.gguf`

Output & embeddings quantized to Q8_0.
All other layers quantized to Q4_K.
Good for CPU inference with limited memory.

`Osmosis-Structure-0.6B-q4_k_s.gguf`

Smallest Q4_K variant, using less memory at the cost of accuracy.
Best for very low-memory setups.

`Osmosis-Structure-0.6B-q6_k.gguf`

Output & embeddings quantized to Q8_0.
All other layers quantized to Q6_K .

`Osmosis-Structure-0.6B-q8_0.gguf`

Fully Q8 quantized model for better accuracy.
Requires more memory but offers higher precision.

`Osmosis-Structure-0.6B-iq3_xs.gguf`

IQ3_XS quantization, optimized for extreme memory efficiency.
Best for ultra-low-memory devices.

`Osmosis-Structure-0.6B-iq3_m.gguf`

IQ3_M quantization, offering a medium block size for better accuracy.
Suitable for low-memory devices.

`Osmosis-Structure-0.6B-q4_0.gguf`

Pure Q4_0 quantization, optimized for ARM devices.
Best for low-memory environments.
Prefer IQ4_NL for better accuracy.

🚀 If you find these models useful

❤ Please click "Like" if you find this useful!

Help me test my AI-Powered Network Monitor Assistant with quantum-ready security checks:
👉 Free Network Monitor

💬 How to test:
Choose an AI assistant type:

TurboLLM (GPT-4o-mini)
HugLLM (Hugginface Open-source)

🟢 TurboLLM – Uses gpt-4o-mini for:

Create custom cmd processors to run .net code on Free Network Monitor Agents
Real-time network diagnostics and monitoring
Security Audits
Penetration testing (Nmap/Metasploit)

🔵 HugLLM – Latest Open-source models:

🌐 Runs on Hugging Face Inference API

💡 Example commands to you could test:

"Give me info on my websites SSL certificate"
"Check if my server is using quantum safe encyption for communication"
"Run a comprehensive security audit on my server"
'"Create a cmd processor to .. (what ever you want)" Note you need to install a Free Network Monitor Agent to run the .net code from. This is a very flexible and powerful feature. Use with caution!

Final Word

I fund the servers used to create these model files, run the Free Network Monitor service, and pay for inference from Novita and OpenAI—all out of my own pocket. All the code behind the model creation and the Free Network Monitor project is open source. Feel free to use whatever you find helpful.

If you appreciate the work, please consider buying me a coffee ☕. Your support helps cover service costs and allows me to raise token limits for everyone.

I'm also open to job opportunities or sponsorship.

Thank you! 😊

`Osmosis-Structure-0.6B`: Small Language Model for Structured Outputs

Osmosis-Structure-0.6B is a specialized small language model (SLM) designed to excel at structured output generation. Despite its compact 0.6B parameter size, this model demonstrates remarkable performance on extracting structured information when paired with supported frameworks.

Our approach leverages structured output during training, forcing our model to only focus on the value for each key declared by the inference engine, which significantly improves the accuracy of the model's ability to produce well-formatted, structured responses across various domains, particularly in mathematical reasoning and problem-solving tasks.

Results

We evaluate the effectiveness of osmosis-enhanced structured generation on challenging mathematical reasoning benchmarks. The following results demonstrate the dramatic performance improvements achieved through structured outputs with osmosis enhancement across different model families - the same technique that powers Osmosis-Structure-0.6B.

Math DAPO 17K Dataset

Model	Structured Output	Structured w/ Osmosis	Performance Gain
Claude 4 Sonnet	15.52%	69.40%	+347%
Claude 4 Opus	15.28%	69.91%	+357%
GPT-4.1	10.53%	70.03%	+565%
OpenAI o3	91.14%	94.05%	+2.9%

Table 1: Performance on Math DAPO 17K.

AIME 1983-2024 Dataset

Model	Structured Output	Structured w/ Osmosis	Performance Gain
Claude 4 Sonnet	16.29%	62.59%	+284%
Claude 4 Opus	22.94%	65.06%	+184%
GPT-4.1	2.79%	39.66%	+1322%
OpenAI o3	92.05%	93.24%	+1.3%

Table 2: Performance on AIME 1983-2024.

Key Insight: These results demonstrate that by allowing models to think freely and leverage test time compute, we are able to increase performance and still maintain the structured guarantee after the fact with a SLM. Osmosis-Structure-0.6B is specifically designed and optimized to maximize these benefits in a compact 0.6B parameter model.

Model Training

Osmosis-Structure-0.6B is built on top of Qwen3-0.6B. We first established a baseline format using 10 samples of randomly generated text and their JSON interpretations. We then applied reinforcement learning to approximately 500,000 examples of JSON-to-natural language pairs, consisting of either reasoning traces with their final outputs, or natural language reports with their expected structured formats.

We used verl as the framework to train our model and SGLang as the rollout backend. To enable structured training, we modified parts of the verl codebase to allow for per sample schema to be passed into the training data.

Usage

SGLang

We recommend an engine like SGLang to be used to serve the model, to serve, run the following:

python3 -m sglang.launch_server --model-path Osmosis/Osmosis-Structure-0.6B --host 0.0.0.0 --api-key osmosis

And to use the endpoint:

import json
from openai import OpenAI

api_key = "osmosis"
api_base_url = "http://0.0.0.0:30000/v1"
client = OpenAI(
    api_key=api_key,
    base_url=api_base_url,
)

# Schema for extracting structured output from reasoning traces
json_schema = json.dumps(
    {
        "type": "object",
        "properties": {
            "answer": {"type": "string"}
        },
        "required": ["answer"]
    }
)

# You can also dump pydantic models to json schema as well

# Example reasoning trace input
reasoning_trace = """
Problem: Solve for x in the equation 2x + 5 = 13

Let me work through this step by step:

First, I need to isolate the term with x. I'll subtract 5 from both sides:
2x + 5 - 5 = 13 - 5
2x = 8

Next, I'll divide both sides by 2 to solve for x:
2x ÷ 2 = 8 ÷ 2
x = 4

Let me verify this answer by substituting back into the original equation:
2(4) + 5 = 8 + 5 = 13 ✓

Ok, which means I got the correct answer, and I'm confident about my answer.
"""
response = client.chat.completions.create(
    model="Osmosis/Osmosis-Structure-0.6B",
    messages=[
        {
            "role": "system",
            "content": f"You are a helpful assistant that understands and translates text to JSON format according to the following schema. {json_schema}"
        },
        {
            "role": "user", 
            "content": reasoning_trace,
        },
    ],
    temperature=0,
    max_tokens=512,
    response_format={
        "type": "json_schema",
        "json_schema": {"name": "reasoning_extraction", "schema": json.loads(json_schema)},
    },
)

print(json.dumps(response.choices[0].message.content, indent=2))

Ollama

You can also use Ollama as an inference provider on local machines, here is a sample code of the setup:

from ollama import chat
from pydantic import BaseModel

class Answer(BaseModel):
  answer: int

reasoning_trace = """
Problem: Solve for x in the equation 2x + 5 = 13

Let me work through this step by step:

First, I need to isolate the term with x. I'll subtract 5 from both sides:
2x + 5 - 5 = 13 - 5
2x = 8

Next, I'll divide both sides by 2 to solve for x:
2x ÷ 2 = 8 ÷ 2
x = 4

Let me verify this answer by substituting back into the original equation:
2(4) + 5 = 8 + 5 = 13 ✓

Ok, which means I got the correct answer, and I'm confident about my answer.
"""

response = chat(
  messages=[
    {
        "role": "system",
        "content": f"You are a helpful assistant that understands and translates text to JSON format according to the following schema. {Answer.model_json_schema()}"
    },
    {
      'role': 'user',
      'content': reasoning_trace,
    }
  ],
  model='Osmosis/Osmosis-Structure-0.6B',
  format=Answer.model_json_schema(),
)

answer = Answer.model_validate_json(response.message.content)
print(answer)

Downloads last month: 506

GGUF

Model size

596M params

Architecture

qwen3

Hardware compatibility

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support