---
license: apache-2.0
datasets:
- PrimeIntellect/fineweb-edu
- PrimeIntellect/fineweb
- PrimeIntellect/StackV1-popular
- mlfoundations/dclm-baseline-1.0-parquet
- open-web-math/open-web-math
- arcee-ai/EvolKit-75K
- arcee-ai/Llama-405B-Logits
- arcee-ai/The-Tomb
- mlabonne/open-perfectblend-fixed
- microsoft/orca-agentinstruct-1M-v1-cleaned
- Post-training-Data-Flywheel/AutoIF-instruct-61k-with-funcs
- Team-ACE/ToolACE
- Synthia-coder
- ServiceNow-AI/M2Lingual
- AI-MO/NuminaMath-TIR
- allenai/tulu-3-sft-personas-code
- allenai/tulu-3-sft-personas-math
- allenai/tulu-3-sft-personas-math-grade
- allenai/tulu-3-sft-personas-algebra
language:
- en
base_model:
- PrimeIntellect/INTELLECT-1-Instruct
base_model_relation: quantized
pipeline_tag: text-generation
quantized_by: bartowski
---

## Exllama v2 Quantizations of INTELLECT-1-Instruct

Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.2.4">turboderp's ExLlamaV2 v0.2.4</a> for quantization.

<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>

Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.

Original model: https://huggingface.co/PrimeIntellect/INTELLECT-1-Instruct

## Prompt format

```
<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>

{prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>


```

## Available sizes

| Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
| ----- | ---- | ------- | ------ | ------ | ------ | ------------ |
| [8_0](https://huggingface.co/bartowski/INTELLECT-1-Instruct-exl2/tree/8_0) | 8.0 | 8.0 | 11.9 GB | 13.3 GB | 15.3 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
| [6_5](https://huggingface.co/bartowski/INTELLECT-1-Instruct-exl2/tree/6_5) | 6.5 | 8.0 | 10.3 GB | 11.7 GB | 13.7 GB | Very similar to 8.0, good tradeoff of size vs performance, **recommended**. |
| [5_0](https://huggingface.co/bartowski/INTELLECT-1-Instruct-exl2/tree/5_0) | 5.0 | 6.0 | 8.3 GB | 9.7 GB | 11.7 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. |
| [4_25](https://huggingface.co/bartowski/INTELLECT-1-Instruct-exl2/tree/4_25) | 4.25 | 6.0 | 7.4 GB | 8.6 GB | 10.6 GB | GPTQ equivalent bits per weight, slightly higher quality. |
| [3_5](https://huggingface.co/bartowski/INTELLECT-1-Instruct-exl2/tree/3_5) | 3.5 | 6.0 | 6.4 GB | 7.8 GB | 9.8 GB | Lower quality, only use if you have to. |

## Download instructions

With git:

```shell
git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/INTELLECT-1-Instruct-exl2 INTELLECT-1-Instruct-exl2-6_5
```

With huggingface hub (credit to TheBloke for instructions):

```shell
pip3 install huggingface-hub
```

To download a specific branch, use the `--revision` parameter. For example, to download the 6.5 bpw branch:

Linux:

```shell
huggingface-cli download bartowski/INTELLECT-1-Instruct-exl2 --revision 6_5 --local-dir INTELLECT-1-Instruct-exl2-6_5
```

Windows (which apparently doesn't like _ in folders sometimes?):

```shell
huggingface-cli download bartowski/INTELLECT-1-Instruct-exl2 --revision 6_5 --local-dir INTELLECT-1-Instruct-exl2-6.5
```

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski