Search public pages, research tools, and SAVRN solutions.
Open-weight model · Text generation
by LG AI Research LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct-AWQ
EXAONE-3.5-7.8B-Instruct-AWQ is an open-weight model for text generation from LG AI Research, released under other. It has 7.8B parameters and a 32,768-token context. At 16-bit it needs about 18.8 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 1M downloads a month.
We introduce EXAONE 3.5, a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.
What it takes to serve EXAONE-3.5-7.8B-Instruct-AWQ (7.8B 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 | 15.6 GB | 18.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 7.8 GB | 9.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 3.9 GB | 4.7 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.
EXAONE-3.5-7.8B-Instruct-AWQ on every accelerator the SAVRN Index prices, at every precision
We introduce EXAONE 3.5, a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research. EXAONE 3.5 language models include: 1) 2.4B model optimized for deployment on small or resource-constrained devices, 2) 7.8B model matching the size of its predecessor but offering improved performance, and 3) 32B model delivering powerful performance. All models support long-context processing of up to 32K tokens. Each model demonstrates state-of-the-art performance in real-world use cases and long-context understanding, while remaining competitive in general domains compared to recently released models of…
Excerpt from the card by LG AI Research, licensed other.
16 files, 5.3 GB in total. The weights are 2 files totalling 5.3 GB in safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00002.safetensors | Weights | 4.5 GB | 1f8706981206 |
| model-00002-of-00002.safetensors | Weights | 838.9 MB | 7d7d1cf66b0b |
| config.json | Configuration | 1.4 KB | — |
| configuration_exaone.py | Configuration | 11.3 KB | — |
| generation_config.json | Configuration | 155 B | — |
| model.safetensors.index.json | Configuration | 61.7 KB | — |
| modeling_exaone.py | Configuration | 24.0 KB | — |
| special_tokens_map.json | Configuration | 563 B | — |
| LICENSE | Documentation | 13.8 KB | — |
| README.md | Documentation | 5.4 KB | — |
| assets/EXAONE_Symbol+BI_3d.png | Other | 249.1 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| merges.txt | Tokenizer | 1.2 MB | — |
| tokenizer.json | Tokenizer | 5.0 MB | — |
| tokenizer_config.json | Tokenizer | 70.7 KB | — |
| vocab.json | Tokenizer | 1.9 MB | — |
Released by LG AI Research through its official repository on Hugging Face.
| Precision | Weights in memory |
|---|---|
| As published | 5.3 GB |
| 16-bit | 15.6 GB |
| 8-bit | 7.8 GB |
| 4-bit | 3.9 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 18.8 GB at 16-bit and 4.7 GB at 4-bit: the weights (7.8B 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.
other, as its publisher declares it. Read the license text before commercial use.
32,768 tokens, from the maximum position embeddings in its published configuration.
Model · Text generation
This model is a fine-tuned version of Jongbin-kr/exaone-7b-verireason-reproduced-1513-fullft-epoch4. It has been trained using TRL. This model was trained with GRPO, a method introduced in DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language 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…
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…
Merged TuneLM checkpoint published from a JarvisLabs GPU job. Do not redistribute Gemma weights outside your licensed use.
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…