# veriloop-coder-e2-nvfp4 by Rodrigo Ramos Da Silveira Source: https://savrn.com/models/veriloop-coder-e2-nvfp4 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve veriloop-coder-e2-nvfp4 (18.3B 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 | 36.7 GB | 44.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 18.3 GB | 22.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 9.2 GB | 11.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 7, 2026. [veriloop-coder-e2-nvfp4 on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/veriloop-coder-e2-nvfp4/gpus) ## Model Card By Rodrigo Ramos Da Silveira, published under apache-2.0, revision 4a489d9380f4. Coding-Optimized Quantized Model (NVIDIA NVFP4) [Original Model ↗](https://huggingface.co/tsinghua-sigs-robot-lab/VeriLoop-E2) · [GitHub](https://github.com/rodrigoramosrs) · Apache-2.0 ### Overview This repository contains an NVFP4 quantization of [VeriLoop E2](https://huggingface.co/tsinghua-sigs-robot-lab/VeriLoop-E2), an open 27B post-trained model built on Qwen3.8-27B for code, mathematics, and physics. Its core reasoning discipline is VeriLoop-Governed Recurrence (VGR): candidate states are recursively proposed, externally checked, and retained only when the protected evidence state improves without regression. Quantized by [Rodrigo Ramos](https://github.com/rodrigoramosrs). ### Quantization Approach Produced with [NVIDIA Model Optimizer](https://github.com/NVIDIA/Model-Optimizer) using the canonical NVFP4_W4A4_WEIGHT_LOCAL_HESSIAN_CFG recipe: static per-block 4-bit weights (group size 16) + dynamic 4-bit activations, FP8 attention, local-Hessian calibration with MSE and fp8 scale sweep. The linear_attn (Gated Delta Net) blocks and the self-attention projections ship in BF16 on purpose — matching validated official NVFP4 releases for this architecture family (which exclude linear_attn* per layer) and the widely-deployed MLP-only NVFP4 pattern. Only the MLP blocks (gate/up/down) are NVFP4. Embeddings, lm_head and small projections stay BF16 per the recipe. [Read the full model card (684 words)](https://savrn.com/models/veriloop-coder-e2-nvfp4/card) ## Configuration Architecture Qwen3_5ForCausalLM Context length (tokens) 262,144 Layers 64 Hidden size 5,120 Feed-forward size 17,408 Attention heads 24 Key/value heads 4 Head dimension 256 Vocabulary size 248,320 Model type qwen3_5_text Quantization modelopt ## Identity and Version Repository rodrigoramosrs/veriloop-coder-e2-nvfp4 Publisher Rodrigo Ramos Da Silveira Task Text generation Modality Text Library transformers Parameters 18.3B parameters Languages en, zh Revision 4a489d9380f4a66d45608a610af87882e1f8145a First published 2026-09-25 Last updated 2026-09-27 ## Files and Weights 25 files, 29.2 GB in total. The weights are 8 files totalling 29.2 GB in safetensors. Weights8 files · 29.2 GB Configuration9 files · 196.9 KB Tokenizer4 files · 22.9 MB Documentation1 file · 6.9 KB Other2 files · 17.7 KB Repository1 file · 1.6 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model-00001-of-00008.safetensors | Weights | 4.0 GB | e3e2f1480bed | | model-00002-of-00008.safetensors | Weights | 4.0 GB | 437be9eda314 | | model-00003-of-00008.safetensors | Weights | 4.0 GB | 5739b61998eb | | model-00004-of-00008.safetensors | Weights | 4.0 GB | b5aafb03e762 | | model-00005-of-00008.safetensors | Weights | 4.0 GB | 6f4c8fb7789f | | model-00006-of-00008.safetensors | Weights | 4.0 GB | c785b3c6b0c8 | | model-00007-of-00008.safetensors | Weights | 2.7 GB | 2f3009f5f490 | | model-00008-of-00008.safetensors | Weights | 2.5 GB | 54d83c1d3663 | | config.json | Configuration | 19.2 KB | — | | configuration.json | Configuration | 29 B | — | | generation_config.json | Configuration | 227 B | — | | hf_quant_config.json | Configuration | 16.2 KB | — | | model.safetensors.index.json | Configuration | 143.2 KB | — | | preprocessor_config.json | Configuration | 390 B | — | | scripts/quantize_veriloop.py | Configuration | 13.4 KB | — | | scripts/sglang-qwen35-loader.patch.py | Configuration | 2.5 KB | — | | scripts/validate_nvfp4.py | Configuration | 1.7 KB | — | | README.md | Documentation | 6.9 KB | — | | chat_template.jinja | Other | 17.5 KB | — | | scripts/Dockerfile.sglang-qwen35fix | Other | 166 B | — | | .gitattributes | Repository | 1.6 KB | — | | merges.txt | Tokenizer | 3.4 MB | — | | tokenizer.json | Tokenizer | 12.8 MB | 0997f410c57a | | tokenizer_config.json | Tokenizer | 18.4 KB | — | | vocab.json | Tokenizer | 6.7 MB | — | ## License and Download License apache-2.0 Access Open weights, no gate Download size 29.2 GB [Download from Rodrigo Ramos Da Silveira](https://huggingface.co/rodrigoramosrs/veriloop-coder-e2-nvfp4) Released by Rodrigo Ramos Da Silveira through its official repository on Hugging Face. [Read the license](https://www.apache.org/licenses/LICENSE-2.0). ## Built From - Derived from tsinghua-sigs-robot-lab/VeriLoop-E2 - Quantized from tsinghua-sigs-robot-lab/VeriLoop-E2 ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 29.2 GB | | 16-bit | 36.7 GB | | 8-bit | 18.3 GB | | 4-bit | 9.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About veriloop-coder-e2-nvfp4 ### How much GPU memory does veriloop-coder-e2-nvfp4 need? About 44 GB at 16-bit and 11 GB at 4-bit: the weights (18.3B parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run veriloop-coder-e2-nvfp4 on? 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. ### Can I use veriloop-coder-e2-nvfp4 commercially? Yes. veriloop-coder-e2-nvfp4 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ### What is veriloop-coder-e2-nvfp4's context length? 262,144 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Text generation ### [NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4](https://savrn.com/models/nvidia-nemotron-3-nano-30b-a3b-nvfp4) [NVIDIA](https://savrn.com/model-publishers/nvidia) September 2025 \- December 2025 The post-training data has a cutoff date of November 28, 2025\. The pre-training data has a cutoff date of June 25, 2025\. Nemotron-Nano-3-30B-A3B-NVFP4 is a quantized version of Nemotron-Nano-3-30B-A3B and is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response. The model's reasoning capabilities can be configured through a flag in the chat template. If the user prefers the model to provide its final answer without intermediate reasoning traces, it can be… Open weights other 18.2B parameters 262,144 tokens transformers [View model](https://savrn.com/models/nvidia-nemotron-3-nano-30b-a3b-nvfp4) Model · Text generation ### [Qwen3.6-35B-A3B-NVFP4](https://savrn.com/models/qwen3-6-35b-a3b-nvfp4) [NVIDIA](https://savrn.com/model-publishers/nvidia) The NVIDIA Qwen3.6-35B-A3B-NVFP4 model is the quantized version of Alibaba's Qwen3.6-35B-A3B model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA Qwen3.6-35B-A3B-NVFP4 model is quantized with Model Optimizer. This model is ready for commercial/non-commercial use. This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party’s requirements for this application and use case; see link to Non-NVIDIA (Qwen3.6-35B-A3B) Model Card from Alibaba. Global Developers looking to take off-the-shelf, pre-quantized models for deployment in AI Agent systems, chatbots… Open weights apache-2.0 18.7B parameters 262,144 tokens Model Optimizer [View model](https://savrn.com/models/qwen3-6-35b-a3b-nvfp4) Model · Text generation ### [NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4](https://savrn.com/models/nvidia-nemotron-3-5-lightning-30b-a3b-nvfp4) [NVIDIA](https://savrn.com/model-publishers/nvidia) The pre-training data has a cutoff date of September 2025. The post-training data has a cutoff date of May 2026. NVIDIA Nemotron™ is a family of open models with open weights, training data, and recipes, delivering leading efficiency and accuracy for building specialized AI agents. NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 is a large language model (LLM) trained by NVIDIA. The model employs a hybrid Mixture-of-Experts architecture, utilizing interleaved Mamba-2 and MoE layers, along with select Attention layers. The Lightning 3.5 model is released alongside a number of speculative decoding methods for faster text generation. The model has 3B active parameters and 30B parameters in total.… Open weights other 17.8B parameters 1,048,576 tokens transformers [View model](https://savrn.com/models/nvidia-nemotron-3-5-lightning-30b-a3b-nvfp4) Model · Text generation ### [K2-Horizon-MoVA-36B-A4B-NSLM-p4mx-q4v](https://savrn.com/models/k2-horizon-mova-36b-a4b-nslm-p4mx-q4v) [Matthew P Barnson](https://savrn.com/model-publishers/txgsync) An experimental NanoSeedLM (SeedLM) compression of IFM/K2-Horizon-MoVA-36B-A4B for Apple Silicon Macs with 32 GB of unified memory. SeedLM keeps each block of 8 weights as a 16-bit seed of a linear feedback shift register (LFSR), 4 coefficients and an exponent. The GPU makes the weights again from the seed at run time. The paper used an FPGA for this. This model uses Metal kernels on the Mac GPU. This is a research proof of concept. It is not a replacement for a calibrated quantization. - Six safetensors shards (22.76 GB in total, at most 4.5 GB each) and model.safetensors.index.json. The other files are config.json, the tokenizer files of the base model and nanoseedlmk2.py, the MLX loader.… Open weights apache-2.0 19.2B parameters 524,288 tokens mlx [View model](https://savrn.com/models/k2-horizon-mova-36b-a4b-nslm-p4mx-q4v) Model · Text generation ### [Ornith-1.5-35B-A3B-NVFP4](https://savrn.com/models/ornith-1-5-35b-a3b-nvfp4) [Ornith](https://savrn.com/model-publishers/ornith-ai) Chirp Chirp! We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement. Ornith-1.5 extends Ornith-1.0 (which was developed on top of Qwen3.5 and Gemma4 with additional continued pretraining, mid-training, and post-training) by expanding the self-improvement loop from scaffold and rollout optimization to jointly optimizing task generation, scaffold construction, and solution rollouts. Rather than relying on a fixed set of human-curated tasks and manually designed harnesses, Ornith-1.5 continuously generates new training tasks, discovers effective strategies for solving them, and improves the policy through reinforcement learning. For… Open weights mit 19.5B parameters 262,144 tokens transformers [View model](https://savrn.com/models/ornith-1-5-35b-a3b-nvfp4) Model · Text generation ### [gpt-neox-20b](https://savrn.com/models/gpt-neox-20b) [EleutherAI](https://savrn.com/model-publishers/eleutherai) GPT-NeoX-20B is a 20 billion parameter autoregressive language model trained on the Pile using the GPT-NeoX library. Its architecture intentionally resembles that of GPT-3, and is almost identical to that of GPT-J- 6B. Its training dataset contains a multitude of English-language texts, reflecting the general-purpose nature of this model. See the accompanying paper for details about model architecture (including how it differs from GPT-3), training procedure, and additional evaluations. Model](https://arxiv.org/abs/2204.06745). For details about the training dataset, see the Pile paper, and its data sheet. Discord](https://discord.gg/zBGx3azzUn), and post them in #release-discussion. Please… Open weights apache-2.0 20.7B parameters 2,048 tokens transformers [View model](https://savrn.com/models/gpt-neox-20b) ## Rodrigo Ramos Da Silveira [All models and datasets](https://savrn.com/model-publishers/rodrigoramosrs) ## Versions - [4a489d9380f4](https://savrn.com/models/veriloop-coder-e2-nvfp4/versions/4a489d9380f4) · current 2026-09-27 ## Explore More - [All text generation models](https://savrn.com/models/tasks/text-generation) - [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-09-27. - [Hugging Face record](https://huggingface.co/rodrigoramosrs/veriloop-coder-e2-nvfp4) - [How the hub is built](https://savrn.com/model-hub/methodology)