AI21’s Jamba Reasoning 3B is a top-performing reasoning model that packs leading scores on intelligence benchmarks and highly-efficient processing into a compact 3B build. Read the full blog post here. The hybrid design combines Transformer attention with Mamba (a state-space model). Mamba layers are more efficient for sequence processing, while attention layers capture complex dependencies. This mix reduces memory overhead, improves throughput, and makes the model run smoothly on laptops, GPUs, and even mobile devices, while maintainig impressive quality. **Smart: Leading intelligence scores** The model outperforms competitors, such as Gemma 3 4B, Llama 3.2 3B, and Granite 4.0 Micro, on a combined intelligence score that averages 6 standard benchmarks.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本→ 输出 文本
Jamba2 3B is an ultra-compact open source model designed to bring enterprise-grade reliability to on-device deployments. At just 3B parameters, it runs efficiently on consumer devices—iPhones, Androids, Macs, and PCs—while maintaining the grounding and instruction-following capabilities required for production use. Released under Apache 2.0 License with a 256K context window, Jamba2 3B enables developers to build reliable AI applications for edge environments. For more details, read the full release blog post. * **On-device deployment:** Runs efficiently on iPhones, Androids, Macs, and PCs * **Ultra-compact footprint:** 3B parameters enabling edge deployments with minimal resources * **Benchmark leadership:** Excels on IFBench, IFEval, Collie, and FACTS * **256K context window:** Processes long documents and knowledge bases * **Apache 2.0 License:** Fully open source for commercial use * **SSM-Transformer architecture:** Memory-efficient design for resource-constrained environments
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本→ 输出 文本
This is a tiny Jamba model used for development, debugging and experimentation over the Jamba architecture. It has 319M parameters (instead of 52B in Jamba 1.5 Mini (and Jamba v0.1) and 398B in Jamba 1.5 Large), and was trained on ~40B tokens. It is great for use in unittests since it is a small model (doesn't take long to download) that has valid and non-random outputs. Yet, **it did not undergo extensive training and should not be expected to generate high-quality text**.
译文排队中 · 当前显示来源原文
分类待补充
输入 待核验→ 输出 待核验
This is a tiny, dummy version of Jamba, used for debugging and experimentation over the Jamba architecture. It has 128M parameters (instead of 52B), **and is initialized with random weights and did not undergo any training.**
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
This is the base version of the Jamba model. We’ve since released a better, instruct-tuned version, Jamba-1.5-Mini. For even greater performance, check out the scaled-up Jamba-1.5-Large. Jamba is a state-of-the-art, hybrid SSM-Transformer LLM. It delivers throughput gains over traditional Transformer-based models, while outperforming or matching the leading models of its size class on most common benchmarks. Jamba is the first production-scale Mamba implementation, which opens up interesting research and application opportunities. While this initial experimentation shows encouraging gains, we expect these to be further enhanced with future optimizations and explorations.
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 音频→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本→ 输出 音频
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本→ 输出 音频
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 音频→ 输出 文本
IndicBART is a multilingual, sequence-to-sequence pre-trained model focusing on Indic languages and English. It currently supports 11 Indian languages and is based on the mBART architecture. You can use IndicBART model to build natural language generation applications for Indian languages by finetuning the model with supervised training data for tasks like machine translation, summarization, question generation, etc. Some salient features of the IndicBART are: Supported languages: Assamese, Bengali, Gujarati, Hindi, Marathi, Odiya, Punjabi, Kannada, Malayalam, Tamil, Telugu and English. Not all of these languages are supported by mBART50 and mT5. The model is much smaller than the mBART and mT5(-base) models, so less computationally expensive for finetuning and decoding. Trained on large Indic language corpora (452 million sentences and 9 billion tokens) which also includes Indian English content. All languages, except English, have been represented in Devanagari script to encourage transfer learning among the related languages. For detailed documentation, look here: https://github.com/AI4Bharat/indic-bart/ and https://indicnlp.ai4bharat.org/indic-bart/
译文排队中 · 当前显示来源原文
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
A multilingual language model trained on IndicCorp v2 and evaluated on IndicXTREME benchmark. The model has 278M parameters and is available in 23 Indic languages and English. The models are trained with various objectives and datasets. The list of models are as follows: - IndicBERT-MLM [Model] - A vanilla BERT style model trained on IndicCorp v2 with the MLM objective - +Samanantar [Model] - TLM as an additional objective with Samanantar Parallel Corpus [Paper] | [Dataset] - +Back-Translation [Model] - TLM as an additional objective by translating the Indic parts of IndicCorp v2 dataset into English w/ IndicTrans model [Model] - IndicBERT-SS [Model] - To encourage better lexical sharing among languages we convert the scripts from Indic languages to Devanagari and train a BERT style model with the MLM objective Fine-tuning scripts are based on transformers library. Create a new conda environment and set it up as follows:
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 音频→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 音频→ 输出 文本
