Model Detail
indic-transcribe-flex
—indic-transcribe-flex is a code generation model with 1B parameters released by bodhan-ai. The model is registered under the automatic-speech-recognition pipeline tag on Hugging Face, distributed under a other license.
indic-transcribe-flex ships with 1B parameters. Total weight footprint is approximately 1.2 GB, which is the relevant figure when planning local-inference VRAM. Access is gated on Hugging Face under the other license, which means a manual approval step before weights can be downloaded.
indic-transcribe-flex is best fit for code completion, repository-scale Q&A, and pair-programming integrations. It is a less obvious choice for one-shot generation of security-critical code without review. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.
Evaluation Awareness Beyond English: Do LLMs Act Differently When Tested in Indic Languages?
arXiv:2610.22333v1 Announce Type: new Abstract: Large language models (LLMs) can often tell when they are being tested. This ability, known as evaluation awareness, can lead models to act more cautiously during safety evaluations than in real use. Existing work has studied evaluation awareness almos
Scaling Indic Pretraining: 2.8 Trillion Tokens and an Open Data Generation Pipeline
bharatSetu: a 1M x 22 bridge for indic translation
arXiv:2610.15705v1 Announce Type: cross Abstract: India's Eighth Schedule recognises 22 languages, yet parallel data is uneven across them, and many language pairs, especially among lower-resource languages such as Bodo, Dogri, Santali and Kashmiri, lack direct bitext. We present BharatSetu 22-Indic
In-Place Indic Tokenizer Expansion and Continual Pretraining of Nemotron-3-Nano-30B-A3B
arXiv:2610.15373v1 Announce Type: cross Abstract: We adapted Nemotron-3-Nano-30B-A3B for eight Indian languages without changing its backbone architecture. We expanded the vocabulary in place from 131,072 to 163,072 entries using continued byte-pair encoding and AdaptMix rebalancing, reducing tokens
A Sovereign, Indic-First, Reasoning Model
arXiv:2610.15371v1 Announce Type: cross Abstract: Gnani-Evon-v3.3-30B-A3B is a long-context hybrid Mamba-Transformer mixture-of-experts model for reasoning, instruction following, and tool use in English and eight Indian languages. Starting from NVIDIA's Nemotron-3-Nano-30B-A3B checkpoint, it retain
Making COMET Comparable Across Scripts: Diagnosis and Correction of Tokeniser-Induced Script Bias in Indic MT Evaluation
arXiv:2610.08159v1 Announce Type: new Abstract: COMET reports translation quality as a single number, and that number is routinely compared across target languages written in different scripts. Such a comparison assumes Script Invariance: the score should not depend on the writing system that carrie
