Open models,
hosted in India.
Every model here is open weight, most are open source, and all run in data centres in India. One API key works for all of them.
- 20+models
- 100%hosted in India
- 40%below list price
All models
MiMo-V2.6 Pro
by XiaomiXiaomi’s flagship model, over 1T parameters, for the most demanding reasoning and agent work.
Chat and reasoning- Context
- 1M
- Input
- Text, image, video, audio
- Licence
- Open source · MIT
- Hosted
- India
₹25 in · ₹50 outper million tokensList price list ₹42 · ₹84GLM-5.3
by Zhipu AIA large reasoning model built for complex software engineering and long-running agent tasks.
Chat and reasoningCoding- Context
- 1M
- Input
- Text
- Licence
- Open weight · own licence
- Hosted
- India
₹81 in · ₹254 outper million tokensList price list ₹135 · ₹424Kimi K3
by Moonshot AIA 2.8T-parameter multimodal reasoning model, strong at coding, knowledge work and agents.
Chat and reasoningCoding- Context
- 1M
- Input
- Text, image, video
- Licence
- Open weight · own licence
- Hosted
- India
₹42 in · ₹752 outper million tokensList price list ₹69 · ₹1,253DeepSeek V4.1 Flash
by DeepSeekA fast mixture-of-experts model, very low cost per token, for high-volume work.
Chat and reasoning- Context
- 1M
- Input
- Text, image
- Licence
- Open source · MIT
- Hosted
- India
₹0.17 in · ₹139 outper million tokensList price list ₹0.29 · ₹231Nemotron 3 Ultra
by NVIDIANVIDIA’s open frontier reasoning model, 550B parameters, built for orchestrating agents.
Chat and reasoningCoding- Context
- 256K
- Input
- Text
- Licence
- Open weight · OpenMDW 1.1
- Hosted
- India
₹29 in · ₹127 outper million tokensList price list ₹48 · ₹212Qwen3.8 27B
by Alibaba (Qwen)A compact vision-language model for coding, research and long-running agent tasks.
Chat and reasoningCoding- Context
- 1M
- Input
- Text, image, video
- Licence
- Open source · Apache 2.0
- Hosted
- India
₹25 in · ₹147 outper million tokensList price list ₹41 · ₹246GLM-5.3 Flash
by Zhipu AIA fast, low-cost GLM for coding and long-running agent tasks.
Chat and reasoningCoding- Context
- 1M
- Input
- Text, image, video
- Licence
- Open source · MIT
- Hosted
- India
₹8.67 in · ₹29 outper million tokensList price list ₹14 · ₹48MiMo-V2.6 Flash
by XiaomiA fast mixture-of-experts model, 309B parameters, for everyday reasoning at scale.
Chat and reasoning- Context
- 1M
- Input
- Text, image, video, audio
- Licence
- Open source · MIT
- Hosted
- India
₹8.09 in · ₹16 outper million tokensList price list ₹13 · ₹27DeepSeek V4 Pro
by DeepSeekDeepSeek’s large model for harder reasoning and coding work.
Chat and reasoningCoding- Context
- 1M
- Input
- Text
- Licence
- Open source · MIT
- Hosted
- India
₹49 in · ₹289 outper million tokensList price list ₹82 · ₹482Qwen3.8 2.4T
by Alibaba (Qwen)The open-weight version of Qwen’s flagship, 2.4T parameters, for the hardest tasks.
Chat and reasoningCoding- Context
- 1M
- Input
- Text
- Licence
- Open weight · own licence
- Hosted
- India
₹116 in · ₹347 outper million tokensList price list ₹193 · ₹578Kimi K2.7 Code
by Moonshot AIBuilt for end-to-end programming tasks over long codebases.
Coding- Context
- 256K
- Input
- Text, image
- Licence
- Open weight · modified MIT
- Hosted
- India
₹39 in · ₹194 outper million tokensList price list ₹65 · ₹323MiniMax M3
by MiniMaxA multimodal model for long-running agents, coding and knowledge work.
Chat and reasoningCoding- Context
- 1M
- Input
- Text, image, video
- Licence
- Open weight · own licence
- Hosted
- India
₹17 in · ₹69 outper million tokensList price list ₹29 · ₹116Gemma 4 31B
by GoogleGoogle’s compact multimodal model with a thinking mode, efficient to run.
Chat and reasoning- Context
- 256K
- Input
- Text, image, video
- Licence
- Open source · Apache 2.0
- Hosted
- India
₹5.20 in · ₹20 outper million tokensList price list ₹8.67 · ₹33Nemotron 3 Super
by NVIDIAAn efficient 120B NVIDIA model for multi-agent applications.
Chat and reasoning- Context
- 256K
- Input
- Text
- Licence
- Open weight · NVIDIA open model licence
- Hosted
- India
₹4.62 in · ₹26 outper million tokensList price list ₹7.71 · ₹43Mistral Small 4
by Mistral AIMistral’s compact model, combining reasoning and vision in one.
Chat and reasoning- Context
- 256K
- Input
- Text, image
- Licence
- Open source · Apache 2.0
- Hosted
- India
₹8.67 in · ₹35 outper million tokensList price list ₹14 · ₹58sarvam-105b
by Sarvam AIBuilt in India for English and 22 Indian languages, from Hindi and Tamil to Santali.
Indian languagesChat and reasoning- Context
- unknown
- Input
- Text
- Licence
- Open source · Apache 2.0
- Hosted
- India
₹18 in · ₹44 outper million tokensList price list ₹29 · ₹73sarvam-30b
by Sarvam AISarvam’s fast mixture-of-experts model for English and 22 Indian languages.
Indian languagesChat and reasoning- Context
- 128K
- Input
- Text
- Licence
- Open source · Apache 2.0
- Hosted
- India
₹5.20 in · ₹20 outper million tokensList price list ₹8.67 · ₹33IndicTrans2
by AI4BharatTranslation between English and 22 Indian languages, and between Indian languages.
Indian languages- Input
- Text
- Licence
- Open source · MIT
- Hosted
- India
₹2.40per 1,000 charactersList price list ₹4.00IndicConformer
by AI4BharatSpeech to text for Indian languages, built for real-world accents and audio.
VoiceIndian languages- Input
- Audio
- Licence
- Open source · MIT
- Hosted
- India
₹18per hour of audioList price list ₹30Indic Parler-TTS
by AI4BharatNatural-sounding text to speech in Indian languages, with controllable voices.
VoiceIndian languages- Input
- Text
- Licence
- Open source · Apache 2.0
- Hosted
- India
₹1.80per 1,000 charactersList price list ₹3.00BGE-M3
by BAAIMultilingual embeddings for search and retrieval across more than 100 languages.
Embeddings- Context
- 8K
- Input
- Text
- Licence
- Open source · MIT
- Hosted
- India
₹0.58 inper million tokensList price list ₹0.96multilingual-e5-large
by MicrosoftCompact multilingual embeddings for search, clustering and classification.
Embeddings- Context
- 512
- Input
- Text
- Licence
- Open source · MIT
- Hosted
- India
₹0.58 inper million tokensList price list ₹0.96
Prices per unit shown, in rupees, 40% below the list price (shown struck through). List prices from public pricing on 4 Oct 2026, converted at ₹96.35 per US$. sarvam-30b is compared with a model of its size, Gemma 4 31B; voice and translation with comparable Sarvam services. Prices exclude GST.
Looking for a model that isn’t here?
Tell us what you need. We add open models as they’re hosted in India.
Start building with sovereign AI.
Open models, hosted in India, behind one API key.