Search public pages, research tools, and SAVRN solutions.
LLaDA-8B-Instruct-PIG is an open-weight model for text generation from IFML, released under MIT License. It has 8B parameters. At 16-bit it needs about 19.2 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 84 downloads a month.
LLaDA-8B-Instruct with Parallel Inconsistency Guidance (PIG), which lets the model commit more tokens per forward pass at the same accuracy. This repository is the whole model: the backbone, a 2.2M-parameter PIG adapter and its calibration tables.
What it takes to serve LLaDA-8B-Instruct-PIG (8B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 16.0 GB | 19.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 8.0 GB | 9.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 4.0 GB | 4.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 18, 2026.
LLaDA-8B-Instruct-PIG on every accelerator the SAVRN Index prices, at every precision
By IFML, published under mit, revision dd2613046500.
LLaDA-8B-Instruct with Parallel Inconsistency Guidance (PIG), which lets the model commit more tokens per forward pass at the same accuracy. This repository is the whole model: the backbone, a 2.2M-parameter PIG adapter and its calibration tables. Paper: PIG: Parallel Inconsistency Guidance for Accelerating Diffusion Language Models, Fruytier, Xue, Shakkottai, Sanghavi (UT Austin), 2026; arXiv:2610.17375. Code: https://github.com/Quentin-Fruytier/PIG-DLM-release Blog: Which tokens can a diffusion language model commit together? Needs transformers 5.x and a GPU with 24 GB. Implementation details, every argument, and how the adapter and tables were fitted: AGENTS.md. MIT. The backbone is…
LLaDA-8B-Instruct with Parallel Inconsistency Guidance (PIG), which lets the model commit more tokens per forward pass at the same accuracy. This repository is the whole model: the backbone, a 2.2M-parameter PIG adapter and its calibration tables.
Paper: PIG: Parallel Inconsistency Guidance for Accelerating Diffusion Language Models, Fruytier, Xue, Shakkottai, Sanghavi (UT Austin), 2026; arXiv:2610.17375. Code: https://github.com/Quentin-Fruytier/PIG-DLM-release Blog: Which tokens can a diffusion language model commit together? Companion: UT-IFML/Nemotron-Labs-Diffusion-3B-PIG
Needs transformers 5.x and a GPU with 24 GB.
from transformers import AutoModel
model = AutoModel.from_pretrained("UT-IFML/LLaDA-8B-Instruct-PIG", trust_remote_code=True)
out = model.generate("Write a Python function that returns the n-th Fibonacci number.", mode="pig.int", phi=0.1)
print(out.text, out.tpf) # tpf: tokens per forward pass
| argument | what it does | default |
|---|---|---|
mode |
"pig.int", "pig.ext", or "confidence" (the Fast-dLLM baseline) |
"pig.int" |
phi |
PIG.INT budget; larger is faster, smaller more accurate (paper range 0.015 to 3) | 0.1 |
theta |
PIG.EXT budget; same direction (paper range 0.125 to 3) | 0.5 |
floor |
PIG.EXT: never commit below this confidence | none |
nu |
baseline: commit everything above this confidence | 0.9 |
geometry, block |
which positions are candidates: "block", "sliding" or "full" |
"block", 32 |
Implementation details, every argument, and how the adapter and tables were fitted: AGENTS.md.
MIT. The backbone is LLaDA-8B-Instruct, released by its authors under the MIT License (NOTICE.md).
@misc{fruytier2026pig,
title = {{PIG}: Parallel Inconsistency Guidance for Accelerating Diffusion Language Models},
author = {Fruytier, Quentin and Xue, Anton and Shakkottai, Sanjay and Sanghavi, Sujay},
year = {2026},
eprint = {2610.17375},
archivePrefix = {arXiv},
primaryClass = {cs.CL},
url = {https://arxiv.org/abs/2610.17375}
}
21 files, 16.0 GB in total. The weights are 7 files totalling 16.0 GB in safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00007.safetensors | Weights | 2.1 GB | addcd50d6dc0 |
| model-00002-of-00007.safetensors | Weights | 2.9 GB | a73ac1a01b3b |
| model-00003-of-00007.safetensors | Weights | 3.0 GB | e69421e0c355 |
| model-00004-of-00007.safetensors | Weights | 3.0 GB | afa61ffc9f95 |
| model-00005-of-00007.safetensors | Weights | 2.9 GB | 90f6c3e4d136 |
| model-00006-of-00007.safetensors | Weights | 2.1 GB | 9358b3a95ffc |
| model-00007-of-00007.safetensors | Weights | 8.7 MB | 354f7f75d65e |
| config.json | Configuration | 2.7 KB | — |
| configuration_llada.py | Configuration | 12.3 KB | — |
| configuration_pig.py | Configuration | 1.9 KB | — |
| model.safetensors.index.json | Configuration | 31.0 KB | — |
| modeling_llada.py | Configuration | 61.9 KB | — |
| modeling_pig.py | Configuration | 41.8 KB | — |
| special_tokens_map.json | Configuration | 747 B | — |
| AGENTS.md | Documentation | 12.4 KB | — |
| LICENSE | Documentation | 1.3 KB | — |
| NOTICE.md | Documentation | 516 B | — |
| README.md | Documentation | 2.7 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| tokenizer.json | Tokenizer | 6.1 MB | — |
| tokenizer_config.json | Tokenizer | 51.7 KB | — |
Released by IFML through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 16.0 GB |
| 16-bit | 16.0 GB |
| 8-bit | 8.0 GB |
| 4-bit | 4.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 19.2 GB at 16-bit and 4.8 GB at 4-bit: the weights (8B parameters) plus a working margin. A long context needs more.
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Yes. LLaDA-8B-Instruct-PIG is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
Full-parameter fine-tuning for TLINK temporal relation classification. mistralai/Ministral-8B-Instruct-2410 AnirbanSaha/tlink-classification-sentence-only The evaluation/ directory contains predictions, the classification report, confusion matrix, and a machine-readable evaluation summary.
The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks. Model Architecture: Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.…
The Model mlx-community/Llama-3.1-8B-Instruct-4bit was converted to MLX format from meta-llama/Llama-3.1-8B-Instruct using mlx-lm version 0.21.4.
Meta developed and released the Meta Llama 3 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety. Model developers Meta Variations Llama 3 comes in two sizes — 8B and 70B parameters — in pre-trained and instruction tuned variants. Input Models input text only. Output Models generate text and code only. Model Architecture Llama 3 is an auto-regressive language…
This model is a fine-tuned version of unsloth/Llama-3.1-8B-unsloth-bnb-4bit. It has been trained using TRL. This model was trained with SFT.
This is the 8B (high-capacity flagship) member of the Med-LLaMA3 family introduced in the paper “Med-LLaMA3: Advancing Medical Question-Answering Through Parameter-Efficient Fine-Tuning of Large Language Models” (Applied Sciences, 2026). The family adapts the LLaMA-3 architecture to the medical domain by training only a small fraction of the base model’s parameters (4.01% for this 8B variant), achieving strong medical question-answering performance while keeping the memory footprint low — enabling development and inference on low-cost, consumer-grade hardware. The 8B variant is the high-capacity model for complex clinical reasoning. It attains a mean accuracy of 75.71% across the eight MMLU…