Merged TuneLM checkpoint published from a JarvisLabs GPU job. Do not redistribute Gemma weights outside your licensed use.
shamrock-0-e4b is an open-weight model for text generation from Oxalis, released under Creative Commons Attribution-NonCommercial 4.0. It has 7.9B parameters and a 131,072-token context. At 16-bit it needs about 19.1 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 1.1k downloads a month.
Gemma 4 E4B fine-tuned for the Oxalis desktop app's language-learning surfaces (camera translate, text translate, tutor, talk, dictionary) across Japanese, Korean, and Simplified Chinese. The E4B sibling As a fair warning...
Parameters7.9B
Context131,072
Weights22.6 GB
Licensecc-by-nc-4.0
AccessOpen weights
Monthly Downloads1.1k
Runs On
What it takes to serve shamrock-0-e4b (7.9B 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 | 15.9 GB | 19.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 7.9 GB | 9.5 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.
shamrock-0-e4b on every accelerator the SAVRN Index prices, at every precision
Model Card
Gemma 4 E4B fine-tuned for the Oxalis desktop app's language-learning surfaces (camera translate, text translate, tutor, talk, dictionary) across Japanese, Korean, and Simplified Chinese. The E4B sibling As a fair warning... this fine-tune will perform poorly outside of the Oxalis app. Feel free to run it for funzies though:) shamrock-0-e4b is licensed under CC BY-NC 4.0 (see LICENSE). You can use, share and adapt it for non-commercial purposes, including personal projects and research, as long as you give credit. Commercial use, such as selling the weights, serving them in a paid product or API, or bundling them into a commercial app, is not permitted. For commercial licensing, reach out…
Excerpt from the card by Oxalis, licensed cc-by-nc-4.0.
Configuration
- Architecture
- Gemma4ForConditionalGeneration
- Context length (tokens)
- 131,072
- Layers
- 42
- Hidden size
- 2,560
- Feed-forward size
- 10,240
- Attention heads
- 8
- Key/value heads
- 2
- Head dimension
- 256
- Vocabulary size
- 262,144
- Sliding window (tokens)
- 512
- Stored precision
- bfloat16
- Model type
- gemma4
Identity and Version
- Repository
- oxalis-ink/shamrock-0-e4b
- Publisher
- Oxalis
- Task
- Text generation
- Modality
- Text
- Library
- transformers
- Parameters
- 7.9B parameters
- Languages
- Not stated by the source
- Revision
- bee58a12bb57f1a2def3231cbec9ba4001be4f1b
- First published
- 2026-08-20
- Last updated
- 2026-10-16
Files and Weights
24 files, 22.7 GB in total. The weights are 8 files totalling 22.6 GB in gguf, safetensors.
Weights8 files · 22.6 GB
Configuration6 files · 223.5 KB
Tokenizer4 files · 64.4 MB
Documentation4 files · 33.5 KB
Other1 file · 18.8 KB
Repository1 file · 133.7 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| MTP/assistant/model.safetensors | Weights | 316.2 MB | ae766a55063b |
| MTP/mtp-shamrock-0-e4b-Q8_0.gguf | Weights | 98.7 MB | 807b51947739 |
| gguf/mmproj-shamrock-0-e4b-f16.gguf | Weights | 990.4 MB | 3ee1f6358ab6 |
| gguf/shamrock-0-e4b-q4_k_m.gguf | Weights | 5.3 GB | 477483323d0c |
| model-00001-of-00004.safetensors | Weights | 5.6 GB | aa3e9d1e0443 |
| model-00002-of-00004.safetensors | Weights | 5.0 GB | 98e3d8b6df44 |
| model-00003-of-00004.safetensors | Weights | 4.9 GB | bfa6b54f0a64 |
| model-00004-of-00004.safetensors | Weights | 389.9 MB | 11f1cd8c8c33 |
| MTP/assistant/config.json | Configuration | 2.4 KB | — |
| MTP/assistant/generation_config.json | Configuration | 304 B | — |
| config.json | Configuration | 6.5 KB | — |
| generation_config.json | Configuration | 204 B | — |
| model.safetensors.index.json | Configuration | 212.4 KB | — |
| processor_config.json | Configuration | 1.7 KB | — |
| LICENSE | Documentation | 19.3 KB | — |
| LICENSE-GEMMA | Documentation | 11.4 KB | — |
| NOTICE | Documentation | 630 B | — |
| README.md | Documentation | 2.2 KB | — |
| chat_template.jinja | Other | 18.8 KB | — |
| .gitattributes | Repository | 133.7 KB | — |
| MTP/assistant/tokenizer.json | Tokenizer | 32.2 MB | 75a6583c1a41 |
| MTP/assistant/tokenizer_config.json | Tokenizer | 1.5 KB | — |
| tokenizer.json | Tokenizer | 32.2 MB | cc8d3a0ce364 |
| tokenizer_config.json | Tokenizer | 26.1 KB | — |
License and Download
- License
- cc-by-nc-4.0
- Access
- Open weights, no gate
- Download size
- 22.6 GB
Download from Oxalis
Released by Oxalis through its official repository on Hugging Face. Read the license.
Built From
- Derived from google/gemma-4-E4B-it
- Quantized from google/gemma-4-E4B-it
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 22.6 GB |
| 16-bit | 15.9 GB |
| 8-bit | 7.9 GB |
| 4-bit | 4.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About shamrock-0-e4b
How much GPU memory does shamrock-0-e4b need?
About 19.1 GB at 16-bit and 4.8 GB at 4-bit: the weights (7.9B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run shamrock-0-e4b 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 shamrock-0-e4b commercially?
Not without separate permission. shamrock-0-e4b is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder.
What is shamrock-0-e4b's context length?
131,072 tokens, from the maximum position embeddings in its published configuration.
Similar Models
Conversational Writing Model｜口语化写作模型 Baihua-Gemma4-E4B 白话 Baihua 是Roland Wayne基于 Google Gemma 4 E4B-it 经过后训练的中文口语化写作模型。把已有内容，或整理好的事实、观点和提纲，写成像人类口述的正文，减少模板化、机械化的表达。 Roland Wayne 是内容创作者，也是白话 Baihua 的训练作者。在这个项目中，他负责确定表达目标、筛选训练材料，并通过实际测试和人工修改评估模型输出。 X：https://x.com/rwayne Github：https://github.com/ozrwayne - 微调方法：LoRA，rank 16，alpha 32；在前一轮风格 adapter 基础上继续进行配对微调。 - 完整权重：BF16 Safetensors，LoRA 已合并，加载完整权重时不需要再叠加同一 adapter。权重文件合计约 15.88 GB；含配套文件的完整包约 15.91 GB（十进制）。 - Ollama 性能测试使用约 14.94 GB 的 BF16 文本 GGUF 运行副本，归一化等少量张量为 FP32；它不包含纯文字推理不需要的音频和图像编码模块。原完整 Safetensors 模型保持独立。 - E4B 是基础模型的官方规格名称，不表示总参数只有 4B。官方标注约 4.5B 有效参数、含嵌入约 8B；当前 Ollama 文本副本实测为 7,463,013,674 个参数，约 7.5B。 - 配置总上下文：131,072…
Peta1 is an autonomous software engineering agent checkpoint based on google/gemma-4-E4B-it (4.3B active parameters), post-trained using a targeted three-stage curriculum: 1. Tree-of-Reasoning (ToR) SFT: Supervised deliberation in Gemma 4's native latent thinking tokens ( thought) enforcing hypothesis formulation, environment simulation, and branch pruning. 2. Capability Frontier Discovery: Screening of candidate tasks (G = 4, T = 0.7) to isolate the 18% informative frontier band (0 < p < 1.0), preventing policy gradient collapse from trivial (36%) and impossible (46%) tasks. 3. Verifiable Outcome GRPO: Policy gradient optimization with verifiable rewards (Rtarget + Rformat + Rtor - lambda…
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…
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.…