Simplify quantization section
Browse files
README.md
CHANGED
|
@@ -8,7 +8,6 @@ base_model:
|
|
| 8 |
tags:
|
| 9 |
- fp8
|
| 10 |
- quantized
|
| 11 |
-
- modelopt
|
| 12 |
- compactifai
|
| 13 |
---
|
| 14 |
|
|
@@ -19,9 +18,9 @@ tags:
|
|
| 19 |
[](TODO_PULSAR_HF_URL)
|
| 20 |
[](https://discord.gg/cGas9uStqp)
|
| 21 |
|
| 22 |
-
Powered by CompactifAI
|
| 23 |
|
| 24 |
-
**Optimized for Fast and Efficient Inference** · **Reduced Memory Footprint** · **FP8
|
| 25 |
|
| 26 |
</div>
|
| 27 |
|
|
@@ -48,7 +47,7 @@ Powered by CompactifAI · Quantized with NVIDIA TensorRT Model Optimizer
|
|
| 48 |
|
| 49 |
**Pulsar 16B FP8** is the **FP8-quantized release** of [Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B), a model based on [NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16), developed by **Multiverse Computing**. The original model is a **~31.6B parameter**, part of the Nemotron model family. It supports **long-context inference up to 1M tokens** and is designed for general-purpose language modeling tasks.
|
| 50 |
|
| 51 |
-
The BF16 release applies **model compression techniques** to significantly reduce parameter count and deployment requirements while maintaining compatibility with the Nemotron Hybrid Mamba2-Transformer with MoE architecture. The resulting model achieves **50% compression**, reducing the parameter count to **16.15B parameters** and lowering memory requirements. **This checkpoint additionally applies FP8
|
| 52 |
|
| 53 |
---
|
| 54 |
|
|
@@ -58,24 +57,18 @@ The BF16 release applies **model compression techniques** to significantly reduc
|
|
| 58 |
|-----------------------|-------------|
|
| 59 |
| Base model | [nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16). **31.6B** total parameters, **3.6B** activated per forward pass (11.34% activation ratio). [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/). |
|
| 60 |
| Pulsar (BF16 release) | [MultiverseComputingCAI/Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B). **16.15B** total parameters, **3.1B** activated per forward pass (19.28% activation ratio) after CompactifAI compression. |
|
| 61 |
-
| Pulsar (this model) | **16.15B** parameters quantized to **FP8**
|
| 62 |
| 📐 **Architecture** | Hybrid Mamba2-Transformer with MoE (same family as the base checkpoint). |
|
| 63 |
| 🛠️ **Tool calling** | Yes. Same tool-call structure and format as [Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16). See [Tool Calling](#tool-calling). |
|
| 64 |
| 🗜️ **Compression** | CompactifAI (proprietary compression technology) |
|
| 65 |
-
| 🔢 **Quantization** | FP8
|
| 66 |
| Primary language | English |
|
| 67 |
|
| 68 |
---
|
| 69 |
|
| 70 |
## Quantization
|
| 71 |
|
| 72 |
-
This checkpoint is a
|
| 73 |
-
|
| 74 |
-
| Field | Value |
|
| 75 |
-
|---------------------|-------|
|
| 76 |
-
| Quantization scheme | **FP8** — weights & activations |
|
| 77 |
-
| Producer | `modelopt` 0.42.0 |
|
| 78 |
-
| Excluded modules | `lm_head`, all Mamba2 `conv1d` modules, and the attention/Mamba projections (`q/k/v/o_proj`, `in_proj`, `out_proj`) of selected layers (kept in higher precision). |
|
| 79 |
|
| 80 |
---
|
| 81 |
|
|
@@ -294,9 +287,9 @@ The base model [nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.
|
|
| 294 |
|
| 295 |
CompactifAI was applied to produce a smaller, efficient model (16B parameters) while aiming to preserve reasoning and tool-use capabilities. Supervised Fine Tuning was applied for improving cabapilities.
|
| 296 |
|
| 297 |
-
### FP8
|
| 298 |
|
| 299 |
-
After compression and fine-tuning, the BF16 checkpoint was quantized to FP8
|
| 300 |
|
| 301 |
---
|
| 302 |
|
|
@@ -417,8 +410,7 @@ Trained mainly on English with added Spanish. No systematic evaluation for langu
|
|
| 417 |
|--------------|--------------------- |
|
| 418 |
| Model name | Pulsar 16B FP8 |
|
| 419 |
| Based on | [MultiverseComputingCAI/Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B) (originally [NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16)) |
|
| 420 |
-
| Precision | FP8 weights
|
| 421 |
-
| Quantization | NVIDIA TensorRT Model Optimizer (`modelopt` 0.42.0) |
|
| 422 |
| Version | v1.5.0 |
|
| 423 |
| Release date | TBD |
|
| 424 |
| Developed by | Multiverse Computing |
|
|
@@ -446,11 +438,11 @@ If you use this model, please cite the base model and pulsar:
|
|
| 446 |
note = {Model developed based on nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 using CompactifAI technology}
|
| 447 |
}
|
| 448 |
@misc{pulsar16bfp8,
|
| 449 |
-
title = {Pulsar 16B FP8: FP8
|
| 450 |
author = {Multiverse Computing},
|
| 451 |
year = {2026},
|
| 452 |
url = {https://huggingface.co/MultiverseComputingCAI/Pulsar-16B-FP8},
|
| 453 |
-
note = {FP8 quantization
|
| 454 |
}
|
| 455 |
```
|
| 456 |
|
|
|
|
| 8 |
tags:
|
| 9 |
- fp8
|
| 10 |
- quantized
|
|
|
|
| 11 |
- compactifai
|
| 12 |
---
|
| 13 |
|
|
|
|
| 18 |
[](TODO_PULSAR_HF_URL)
|
| 19 |
[](https://discord.gg/cGas9uStqp)
|
| 20 |
|
| 21 |
+
Powered by CompactifAI
|
| 22 |
|
| 23 |
+
**Optimized for Fast and Efficient Inference** · **Reduced Memory Footprint** · **FP8 Quantization**
|
| 24 |
|
| 25 |
</div>
|
| 26 |
|
|
|
|
| 47 |
|
| 48 |
**Pulsar 16B FP8** is the **FP8-quantized release** of [Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B), a model based on [NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16), developed by **Multiverse Computing**. The original model is a **~31.6B parameter**, part of the Nemotron model family. It supports **long-context inference up to 1M tokens** and is designed for general-purpose language modeling tasks.
|
| 49 |
|
| 50 |
+
The BF16 release applies **model compression techniques** to significantly reduce parameter count and deployment requirements while maintaining compatibility with the Nemotron Hybrid Mamba2-Transformer with MoE architecture. The resulting model achieves **50% compression**, reducing the parameter count to **16.15B parameters** and lowering memory requirements. **This checkpoint additionally applies FP8 quantization** on top of the compressed model.
|
| 51 |
|
| 52 |
---
|
| 53 |
|
|
|
|
| 57 |
|-----------------------|-------------|
|
| 58 |
| Base model | [nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16). **31.6B** total parameters, **3.6B** activated per forward pass (11.34% activation ratio). [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/). |
|
| 59 |
| Pulsar (BF16 release) | [MultiverseComputingCAI/Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B). **16.15B** total parameters, **3.1B** activated per forward pass (19.28% activation ratio) after CompactifAI compression. |
|
| 60 |
+
| Pulsar (this model) | **16.15B** parameters with weights quantized to **FP8**. |
|
| 61 |
| 📐 **Architecture** | Hybrid Mamba2-Transformer with MoE (same family as the base checkpoint). |
|
| 62 |
| 🛠️ **Tool calling** | Yes. Same tool-call structure and format as [Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16). See [Tool Calling](#tool-calling). |
|
| 63 |
| 🗜️ **Compression** | CompactifAI (proprietary compression technology) |
|
| 64 |
+
| 🔢 **Quantization** | FP8 (weights). |
|
| 65 |
| Primary language | English |
|
| 66 |
|
| 67 |
---
|
| 68 |
|
| 69 |
## Quantization
|
| 70 |
|
| 71 |
+
This checkpoint is a quantized version of [Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B): its weights have been quantized to **FP8**.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
|
| 73 |
---
|
| 74 |
|
|
|
|
| 287 |
|
| 288 |
CompactifAI was applied to produce a smaller, efficient model (16B parameters) while aiming to preserve reasoning and tool-use capabilities. Supervised Fine Tuning was applied for improving cabapilities.
|
| 289 |
|
| 290 |
+
### FP8 Quantization
|
| 291 |
|
| 292 |
+
After compression and fine-tuning, the BF16 checkpoint was quantized to FP8.
|
| 293 |
|
| 294 |
---
|
| 295 |
|
|
|
|
| 410 |
|--------------|--------------------- |
|
| 411 |
| Model name | Pulsar 16B FP8 |
|
| 412 |
| Based on | [MultiverseComputingCAI/Pulsar-16B](https://huggingface.co/MultiverseComputingCAI/Pulsar-16B) (originally [NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16)) |
|
| 413 |
+
| Precision | FP8 weights |
|
|
|
|
| 414 |
| Version | v1.5.0 |
|
| 415 |
| Release date | TBD |
|
| 416 |
| Developed by | Multiverse Computing |
|
|
|
|
| 438 |
note = {Model developed based on nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 using CompactifAI technology}
|
| 439 |
}
|
| 440 |
@misc{pulsar16bfp8,
|
| 441 |
+
title = {Pulsar 16B FP8: FP8-quantized release of Pulsar-16B},
|
| 442 |
author = {Multiverse Computing},
|
| 443 |
year = {2026},
|
| 444 |
url = {https://huggingface.co/MultiverseComputingCAI/Pulsar-16B-FP8},
|
| 445 |
+
note = {FP8 weight quantization of Pulsar-16B}
|
| 446 |
}
|
| 447 |
```
|
| 448 |
|