DISCOVER

Your next building block.

Models, datasets and tools from their original sources. Search the catalog, inspect the evidence, and find a practical place to start.

Filter resources

Results update as you select.

Indian sources

Publishers with verified Indian provenance. Separate from language support.

Resource typePrimary taskLanguageLicenseSource platformMetadata statusSource updated sinceMinimum parametersMaximum parametersActual counts, when source metadata supplies them.ModalityIndian-language relevanceLocal deployment indicatedOpen weights indicated

Reset filters

Filters use attributed source information. Unknown publisher origins and capabilities stay unknown.

4 resourcesSort: Compare

evaluation

AI4Bharat

indic-llm-eval-suite

Indian-language evaluation tooling from AI4Bharat. Compare results only when task versions and evaluation settings agree.

EvaluationIndic languageIndian sourceIndian languages

GitHubLicense unknown

evaluation

UKGovernmentBEIS

inspect_ai

Inspect: A framework for large language model evaluations

EvaluationAgents

GitHubMIT

evaluation

EleutherAI

lm-evaluation-harness

A framework for few-shot evaluation of language models.

EvaluationReproducibility

GitHubMIT

evaluation

promptfoo

Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, DeepSeek, and more. Simple declarative configs with command line and CI/CD integration. Used by OpenAI and Anthropic.

EvaluationRed teaming

GitHubMIT

Recommended order prioritizes verified Indian publishers, followed by other sources, with verified China-based publishers later. Relevance and freshness order each group. This is an editorial preference, not a quality score. Your filters stay in the shareable URL.

Your next building block.

Catalog results

One moment.

Your next building block.

Catalog results