README.md

---
license: apache-2.0
base_model:
- mistralai/Mistral-7B-Instruct-v0.3
pipeline_tag: text2text-generation
---

# Elastic model: Mistral-7B-Instruct-v0.3. Fastest and most flexible models for self-serving.

Elastic models are the models produced by TheStage AI ANNA: Automated Neural Networks Accelerator. ANNA allows you to control model size, latency and quality with a simple slider movement. For each model, ANNA produces a series of optimized models:

* __XL__: Mathematically equivalent neural network, optimized with our DNN compiler. 

* __L__: Near lossless model, with less than 1% degradation obtained on corresponding benchmarks.

* __M__: Faster model, with accuracy degradation less than 1.5%.

* __S__: The fastest model, with accuracy degradation less than 2%.


__Goals of elastic models:__

* Provide flexibility in cost vs quality selection for inference
* Provide clear quality and latency benchmarks
* Provide interface of HF libraries: transformers and diffusers with a single line of code
* Provide models supported on a wide range of hardware, which are pre-compiled and require no JIT.
* Provide the best models and service for self-hosting.

> It's important to note that specific quality degradation can vary from model to model. For instance, with an S model, you can have 0.5% degradation as well.


![image/png](https://cdn-uploads.huggingface.co/production/uploads/6487003ecd55eec571d14f96/BV-6-DVgNW-aqcmY1Zr20.png)

-----

## Inference

To infer our models, you just need to replace `transformers` import with `elastic_models.transformers`:

```python
import torch
from transformers import AutoTokenizer
from elastic_models.transformers import AutoModelForCausalLM

# Currently we require to have your HF token
# as we use original weights for part of layers and
# model confugaration as well
model_name = "mistralai/Mistral-7B-Instruct-v0.3"
hf_token = ''
device = torch.device("cuda")

# Create mode
tokenizer = AutoTokenizer.from_pretrained(
    model_name, token=hf_token
)
model = AutoModelForCausalLM.from_pretrained(
    model_name, 
    token=hf_token,
    torch_dtype=torch.bfloat16,
    attn_implementation="sdpa",
    mode='s'
).to(device)
model.generation_config.pad_token_id = tokenizer.eos_token_id

# Inference simple as transformers library
prompt = "Describe basics of DNNs quantization."
inputs = tokenizer(prompt, return_tensors="pt")
inputs.to(device)

with torch.inference_mode:
    generate_ids = model.generate(**inputs, max_length=500)

input_len = inputs['input_ids'].shape[1]
generate_ids = generate_ids[:, input_len:]
output = tokenizer.batch_decode(
    generate_ids,
    skip_special_tokens=True, 
    clean_up_tokenization_spaces=False
)[0]

# Validate answer
print(f"# Q:\n{prompt}\n")
print(f"# A:\n{output}\n")
```

__System requirements:__
* GPUs: H100, L40s
* CPU: AMD, Intel
* Python: 3.10-3.12


To work with our models just run these lines in your terminal:

```shell
pip install thestage
pip install elastic_models==0.0.3\
 --index-url https://thestage.jfrog.io/artifactory/api/pypi/pypi-thestage-ai-production/simple\
 --extra-index-url https://pypi.nvidia.com\
 --extra-index-url https://pypi.org/simple

pip install flash_attn==2.7.3 --no-build-isolation
pip uninstall apex
echo "{
    "meta-llama/Llama-3.2-1B-Instruct": 6,
    "mistralai/Mistral-7B-Instruct-v0.3": 7,
    "black-forest-labs/FLUX.1-schnell": 1,
    "black-forest-labs/FLUX.1-dev": 5
}" > model_name_id.json
export ELASTIC_MODEL_ID_MAPPING=./model_name_id.json
```

Then go to [app.thestage.ai](https://app.thestage.ai), login and generate API token from your profile page. Set up API token as follows:

```shell
thestage config set --api-token <YOUR_API_TOKEN>
```

Congrats, now you can use accelerated models!

----

## Benchmarks

Benchmarking is one of the most important procedures during model acceleration. We aim to provide clear performance metrics for models using our algorithms. The `W8A8, int8 column` indicates that we applied W8A8 quantization with int8 data type to all linear layers and used the same calibration data as for ANNA. The S model achieves practically identical speed but much higher quality, as ANNA knows how to improve quantization quality on sensitive layers!

### Quality benchmarks

<!-- For quality evaluation we have used: #TODO link to github -->

| Metric/Model  | S | M | L | XL | Original | W8A8, int8 |
|---------------|---|---|---|----|----------|------------|
| MMLU          | 59.7 | 60.1 |  60.8 | 61.4  | 61.4        | 28          |
| PIQA          | 80.8 | 82   | 81.7  | 81.5  | 81.5        | 65.3          |
| Arc Challenge | 56.6 | 55.1 | 56.8  | 57.4  | 57.4        | 33.2         |
| Winogrande    | 73.2 | 72.3 | 73.2  | 74.1  | 74.1        | 57          |


* **MMLU**:Evaluates general knowledge across 57 subjects including science, humanities, engineering, and more. Shows model's ability to handle diverse academic topics.
* **PIQA**: Evaluates physical commonsense reasoning through questions about everyday physical interactions. Shows model's understanding of real-world physics concepts.
* **Arc Challenge**: Evaluates grade-school level multiple-choice questions requiring reasoning. Shows model's ability to solve complex reasoning tasks.
* **Winogrande**: Evaluates commonsense reasoning through sentence completion tasks. Shows model's capability to understand context and resolve ambiguity.

### Latency benchmarks

__100 input/300 output; tok/s:__

| GPU/Model | S   | M | L | XL | Original | W8A8, int8 |
|-----------|-----|---|---|----|----------|------------|
| H100      | 189 | 166 | 148 | 134  | 49       | 192          |
| L40s      | 79  | 68 | 59 | 47  | 38    | 82         |



## Links

* __Platform__: [app.thestage.ai](app.thestage.ai)
<!-- * __Elastic models Github__: [app.thestage.ai](app.thestage.ai) -->
* __Subscribe for updates__: [TheStageAI X](https://x.com/TheStageAI)
* __Contact email__: contact@thestage.ai