IndicStack
Platform
ConsultingModelsAgentsComplianceBlog
Request Early Access
IndicStack

Product

  • Platform
  • Models
  • Agents
  • Consulting

Solutions

  • IT Services & Agencies
  • SaaS Startups
  • Regulated Industries
  • WhatsApp Automation

Company

  • About
  • Compliance
  • Blog

Legal

  • Privacy Policy
  • Terms of Service

IndicStack Consultancy Services LLP — Built for Indian AI builders.

18 models across 5 categories

Model Catalog

Indic-native and multilingual models hosted on Indian infrastructure. All accessible via OpenAI-compatible API.

Sarvam-M

24Bctx 32KChat

Instruction-tuned for 10 Indian languages. Strong for regional customer support and voice agent backends.

Indic-nativeDefaultApache 2.0
hibntatekngumrmlorpaen

Sarvam-30B

32.2B (MoE)ctx 8KChat

Flagship Indic chat model with extensive code-mixing support across 22 Indian languages.

Indic-nativeDefaultApache 2.0
hibntatekngumrmlpaorasen+10

Qwen3-8B

8Bctx 128KChat

High-throughput multilingual chat. Best economy option for English-first workloads with broad Indic coverage.

EconomyApache 2.0
enhi+100 languages

Gemma 4 E4B

8B (4.5B eff.)ctx 128KChat

Google's latest multimodal model. Low-cost serving with native support for major Indian languages.

EconomyApache 2.0
enhibntateknml+133 languages

DeepSeek-R1 Distill 14B

14Bctx 128KChat

R1 chain-of-thought reasoning distilled from Qwen2.5-14B. Fits single A100 comfortably at FP16.

DefaultMIT
enzhmultilingual

Gemma 4 26B-A4B

25.2B (3.8B active)ctx 256KChat

MoE design delivers frontier-quality reasoning at economy-tier throughput cost. 256K context window.

DefaultApache 2.0
enhibntateknml+133 languages

Qwen3-32B

32Bctx 128KChat

Strong coding and reasoning with hybrid thinking mode. High-context agentic and RAG workloads.

DefaultApache 2.0
enhi+100 languages

DeepSeek-R1 Distill 32B

32Bctx 128KChat

Best open reasoning model under 70B. Outperforms o1-mini on AIME 2024, MATH-500, and LiveCodeBench.

DefaultMIT
enzhmultilingual

Sarvam-105B

106Bctx 8KChat

Highest-quality Indic model available. Dedicated GPU deployment — contact us to discuss fit.

Indic-nativePremiumApache 2.0
hibntatekngumrmlpaorasen+10

IndicConformer

600MSpeech-to-Text

Fast, accurate ASR across all 22 scheduled Indian languages including low-resource variants.

Indic-nativeEconomyMIT
22 Indian languages

Whisper Large V3 (Hindi)

1.5BSpeech-to-Text

Whisper V3 fine-tuned on Vaani Hindi dataset. Handles diverse accents, noisy audio, and code-mixed speech.

EconomyApache 2.0
hi

Qwen3-ASR-1.7B

1.7BSpeech-to-Text

State-of-the-art open ASR with unified offline and streaming inference. Language identification included.

EconomyApache 2.0
hi+30 languages

Indic Parler TTS

937MText-to-Speech

Expressive TTS with 500+ speaker voices across 18 Indian languages. Prompt-controlled voice style.

Indic-nativeEconomyApache 2.0
18 Indian languages

Indic-Mio

0.6BText-to-Speech

44kHz TTS with zero-shot voice cloning and code-mixed text support across all 22 scheduled Indian languages.

Indic-nativeEconomyApache 2.0
22 Indian languages

Qwen3-Embedding-0.6B

0.6Bctx 32KEmbeddings

MTEB-leading multilingual embeddings. Best for RAG over Indic and multilingual document corpora.

EconomyApache 2.0
119 languages

Granite Embedding 311M

311Mctx 32KEmbeddings

IBM's multilingual retrieval model with Matryoshka truncation and ONNX export. Minimal serving overhead.

EconomyApache 2.0
200+ languages

Qwen3-Reranker-0.6B

0.6Bctx 32KReranker

Lightweight cross-encoder reranker for improving retrieval precision in multilingual RAG pipelines.

EconomyApache 2.0
multilingual

GTE Multilingual Reranker

306Mctx 8KReranker

Encoder-only reranker with ~10x throughput advantage over decoder-based alternatives. Proven in production.

EconomyApache 2.0
70+ languages

Need a specific model?

We evaluate new models continuously. Tell us what you need and we will check if it fits our serving infrastructure.

Request a Model