Picture three very different people trying to use AI in 2026. A farmer in rural Maharashtra asking about crop insurance in Marathi. A patient in Tamil Nadu trying to understand a prescription written in medical English. A field worker filling out a government form on a ₹7,000 phone with patchy 2G signal.

For all three, the giant AI models making headlines — the ones with hundreds of billions of parameters, running in data centres thousands of kilometres away — are the wrong tool. Not because they aren’t smart. Because they’re expensive, slow on a bad connection, awkward when the data is sensitive, and often mediocre in the language the person is actually speaking.

This is the argument at the heart of a chapter in EY India’s The AIdea of India: Outlook 2026 report: for a country as linguistically diverse and digitally uneven as India, small language models aren’t a cheaper compromise. They’re the actual right tool for the job — and a genuine opportunity to lead, not just catch up.

First, What Is a “Small” Language Model, Really?

You’ve probably heard of large language models — the AI behind tools like ChatGPT or Claude, built with hundreds of billions of parameters (think of parameters as the tiny adjustable dials inside a model that let it learn patterns; more dials generally means more capability, but also more cost).

A small language model (SLM) is the same basic idea, dramatically scaled down — typically between 1 billion and 15 billion parameters as of 2026. That might still sound huge, but compare it to frontier models running past 100 billion parameters, and you start to see the gap.

The surprising part is how well these smaller models now perform. A few clever techniques make this possible:

• Learning from a teacher. A large, expensive model “teaches” a smaller one, passing down its knowledge in a distilled form rather than making the small model learn everything from scratch.
• Better data instead of more data. Some of the best small models today were trained on carefully chosen, high-quality data rather than a giant, messy scrape of the internet — a bit like the difference between studying a well-organised textbook versus every scrap of paper you can find.
• Compression. Techniques that shrink a model’s memory footprint by roughly half without meaningfully hurting its quality, the AI equivalent of zipping a large file.
• Only waking up the parts you need. Some models technically contain tens of billions of parameters but only activate a small slice of them for any given question — getting much of the power of a large model while using a fraction of the resources.

The result of all this: a compact model released in early 2025 can now outperform a model seven times its size on standard reasoning tests. Two years ago, that level of performance needed a much, much bigger model.

Small vs. Big: The Honest Trade-Off

Neither type of model is simply “better.” They’re built for different jobs.

The practical rule most teams are landing on: use a small model for the repetitive, high-volume 80% of everyday requests, and reserve the big, expensive model for the genuinely hard 20% that actually needs deep reasoning.

Why the Economics Suddenly Changed

Three separate trends collided to make this shift possible right now, not five years from now:

AI got dramatically cheaper to run. The cost of running a mid-tier AI model has fallen by roughly 280 times in just two years, largely thanks to smaller, smarter models doing more with less.

The hardware caught up. Analysts expect over half of all new PCs sold in 2026 to be capable of running AI models directly on the device, no internet connection required.

The way we use AI changed. A growing share of real-world AI use isn’t a human having an open-ended conversation — it’s an AI “agent” doing the same narrow task over and over: extracting an invoice number, sorting a support ticket, translating a phrase. That kind of repetitive, specialised work is exactly what a small model is built for. Using a massive, expensive model to extract an invoice number 40,000 times a day is, frankly, overkill.

One industry forecast puts it plainly: by 2027, businesses are expected to use small, task-specific AI models three times more often than big general-purpose ones.

Why This Is Especially India’s Opportunity

The case for small models isn’t unique to India, but the fit is unusually strong here, for a few concrete reasons.

Language. India officially recognises 22 languages, and everyday life involves many more. The big global AI models are excellent in English, decent in a handful of major world languages, and genuinely weak in languages like Bhojpuri, Maithili, or Santali. Training a smaller, focused model on a specific Indian language is far more achievable than trying to make a giant global model equally fluent in all of them.

Infrastructure. Slow networks and budget smartphones are simply everyday reality for a huge share of India’s population. A small model that works offline on an entry-level phone isn’t a lesser version of AI — for millions of people, it’s the only version that actually works at all.

Data rules and cost control. Because small models can run on private servers or directly on a device, sensitive data never has to leave the country, or even leave the building. That matters enormously for banks, hospitals, and government services bound by data protection law, and it also makes monthly costs far more predictable.

A track record of frugal engineering. India has already solved population-scale problems on a budget before — think of Aadhaar (the world’s largest biometric ID system) and UPI (India’s now-massive digital payments network). Both were built around the same philosophy: solve the real problem cheaply, at enormous scale. Small language models fit that same mindset almost perfectly.

To be fair, EY’s report is honest about the flip side too: India’s digital content in many regional languages is still thin, high-end computing power remains limited and expensive, and the research ecosystem is still maturing. Small models aren’t a way to pretend those challenges don’t exist — they’re a way to build something genuinely useful despite them.

What’s Actually Happening in India Right Now

This isn’t a future possibility — it’s already underway. India’s government-backed IndiaAI Mission has funded twenty home-grown AI model projects, a mix of large and small models, across a dozen organisations. Several have already launched:

• A model that handles real-time conversation and reasoning, trained from scratch entirely within India, with its underlying technology released publicly for others to build on.
• A model specifically built for governance, agriculture, health and education, covering all 22 scheduled languages.
• A voice-cloning system that can convincingly reproduce a voice in twelve Indian languages from under ten seconds of sample audio, designed specifically to work well on low-bandwidth connections.

These aren’t just research demos. India’s national identity system has already integrated one of these models into fully offline, on-premise voice services in ten languages. A major insurer is rolling out a similar system to 80 million customers. A national translation platform now sits quietly behind government portals used by well over 100 million people.

Where Small Models Are Actually Useful

This is the part that matters most for anyone running a business, not just AI researchers. Small language models are already being put to work in genuinely practical ways:

• Banking and insurance: customer service in regional languages, reading and sorting scanned documents, flagging potential fraud, and keeping sensitive financial data entirely in-house.
• Healthcare: explaining a prescription to a patient in their own language, summarising discharge notes, and running fully offline on a health worker’s phone in areas with no reliable internet.
• Agriculture: crop advice by voice in a farmer’s own language, identifying pests or disease from a photo taken directly on a phone, no upload required.
• Government services: checking eligibility for a welfare scheme, sorting citizen complaints, and summarising legal or land documents — all in a way that can be fully audited and kept within the country.
• Everyday business software: sorting IT support tickets, reviewing code without sensitive company data ever leaving the building, and pulling structured information out of meeting notes or customer records.
• Retail: organising product catalogues at scale, summarising customer reviews, and running in-store kiosks that respond instantly without needing a live internet connection.

The Smartest Approach Isn’t Choosing One or the Other

The most useful conclusion in EY’s report isn’t “small models win.” It’s that India’s real advantage comes from combining both intelligently: small models handling the huge volume of routine, everyday requests, and large models stepping in only when a question genuinely needs deep, open-ended reasoning.

In practice, this looks like a simple routing system: an easy request gets handled instantly by a small, specialised model running locally. A request in a specific regional language gets routed to a model built for that language. Only the genuinely difficult, unusual questions get sent to an expensive, powerful model in the cloud — and even then, with any sensitive personal information stripped out first.

Most companies that try AI and find it disappointingly expensive made one specific mistake: they sent every single request to the big, expensive model, and only discovered the bill afterward.

What to Watch Out For

It’s worth being honest about the limits here too. Small models are narrow by design — a model fine-tuned to sort insurance claims will confidently give a wrong answer if you ask it something outside that lane, so knowing exactly what a model is meant to do (and not letting it stray outside that) really matters.

Independently verifying that these models actually perform as claimed, especially in less commonly represented languages, is still a developing area in India — and it matters, because real decisions and real government spending increasingly rest on those performance numbers. And while these models can run on Indian soil, most of the underlying computer chips they run on still come from abroad — a reminder that owning the model is not quite the same as owning the entire supply chain behind it.

The Bigger Picture

India was never likely to out-spend the world’s biggest AI labs in a race to build the single largest model. That was never really the game worth playing.

The stronger position — the one India has already proven it can win, with Aadhaar and UPI — is building the cheapest, most inclusive, most genuinely usable version of a technology, at a scale few other countries can match, and then sharing that model with the rest of the world. Small language models are simply the AI version of that same successful playbook.

Small, in India’s case, isn’t a limitation. It’s the strategy.

Frequently Asked Questions

What is a small language model?
A small language model (SLM) is a compact AI model, typically between 1 and 15 billion parameters, designed to run efficiently on limited hardware — including phones and laptops — while still handling specific language, reasoning or coding tasks well.

How is an SLM different from a large language model like ChatGPT or Claude?
Large language models are far bigger and more broadly capable, but are expensive to run, need an internet connection to a data centre, and are slower to respond. Small language models are cheaper, faster, can often run offline or on-device, and excel at narrow, specific tasks rather than open-ended general conversation.

Why are small language models especially relevant for India?
India’s linguistic diversity, uneven internet infrastructure, low-cost device usage, and data privacy requirements all favour smaller, locally-deployable models over large cloud-based ones. Small models can be trained for specific Indian languages, work offline, and keep sensitive data within the country.

Are small language models less accurate than large ones?
For narrow, specific tasks they’re trained for, small models can match or even outperform much larger models. For broad, open-ended reasoning across many topics, large models still generally have the edge. Most real-world systems use a mix of both.

What is the IndiaAI Mission?
The IndiaAI Mission is a government-backed initiative funding the development of home-grown AI models, including both large and small language models, built by Indian research organisations and companies, with several models already publicly released.

Share

The post Small Language Models: How India Wins AI Without Winning the Parameter Race appeared first on Blog.

Here’s what actually happened, what makes it credible rather than hype, and why the “closed AI has a permanent moat” argument just got harder to make with a straight face.

What Moonshot AI Actually Shipped

Kimi K3 is Moonshot AI’s new flagship model — 2.8 trillion parameters, built as a Mixture-of-Experts (MoE) system. The key detail that makes that number less intimidating than it sounds: only 16 of its 896 “experts” activate for any given token. That architecture is what lets a model with genuinely massive parameter count keep real inference cost far below what a dense model of the same size would demand.

Moonshot is calling it the first “open 3T-class” model, edging past DeepSeek’s 1.6T V4 Pro for that particular title. Alongside the scale, K3 ships with a 1-million-token context window, native vision support, and a new attention mechanism — Kimi Delta Attention (KDA) — that the team reports delivers up to roughly 6x faster decoding at long context lengths. Most models slow down noticeably as context grows; KDA is specifically engineered to fight that.

The headline result: on Arena’s Frontend Code leaderboard — where real developers vote on blind, head-to-head matchups between model outputs — K3 landed at #1. That’s a 17-place jump over its own predecessor, Kimi K2.6, and it placed ahead of Anthropic’s Claude Fable 5 in that specific comparison.

The Part That Makes This Credible, Not Hype

Here’s the detail that separates this from the usual “new model claims to beat everything” launch post: Moonshot’s own materials openly admit that K3 trails both Claude Fable 5 and GPT-5.6 Sol on general intelligence benchmarks.

It is not, by Moonshot’s own account, the smartest model available. What it won was one specific, real-world-relevant contest — front-end code that actual developers preferred, in blind evaluation. That’s a narrower claim than “best AI model,” and a far more believable one.

The moat closed AI labs have isn’t “we’re smarter.” It’s “we’re smarter at everything, all the time.” Kimi K3 just showed that gap can close fast, on the tasks people actually care about.

Framed that way, this isn’t really a “China beats America” story, even though that’s the framing a lot of coverage has reached for. It’s a more precise and more interesting one: an open model caught up on a specific task that matters commercially, while still openly trailing on general capability. That’s exactly the kind of gap that tends to narrow over successive model generations rather than widen — which is the actual reason this is worth paying attention to.

What Makes K3 Technically Different

Two things are doing the real work under the hood, beyond the raw parameter count:

• Kimi Delta Attention (KDA). A hybrid linear attention mechanism built specifically to keep long-context performance from degrading. At a 1-million-token context window, this matters a lot in practice — it’s the difference between a model that’s usable on a large real codebase and one that’s only fast on toy examples.
• Sparse Mixture-of-Experts routing. Activating only 16 of 896 experts per token is what makes a 2.8T-parameter model economically viable to serve at all. This is the same broad architectural family used by other recent frontier models, but Moonshot has pushed the expert count and routing design further than most public releases.

Pricing

K3 is priced at roughly $0.30 per million tokens for cache-hit input, up to $3 per million tokens for cache-miss input, and $15 per million tokens for output. That’s meaningfully cheaper than most closed frontier models — not free, but a real and relevant gap for any team weighing cost against capability.

How to Actually Use It

• Via API. Model ID kimi-k3, available through Moonshot’s own platform in an OpenAI-compatible API format — low-friction to try if your stack already calls GPT-style endpoints.
• Via the Kimi app. Available directly through kimi.com for casual, non-developer use.
• Self-hosting. Not yet available. Full open weights were promised by July 27, 2026; until that lands, access is limited to the hosted API.
• Practical recommendation from early adopters. Several development teams evaluating K3 are treating it as a candidate default-plus-fallback model rather than a wholesale production swap — trialling it against real diffs, tests, and builds before committing meaningful traffic to it.

One more data point worth including for balance: Moonshot briefly paused new K3 subscriptions on July 20, citing GPU capacity limits after unexpectedly high demand. That’s a good signal of genuine interest in the model, and also a fair signal that the infrastructure behind it is still catching up to that demand.

Why This Matters If You’re Building With AI

The practical takeaway isn’t “switch to Kimi K3 today.” It’s a reminder about how quickly the gap between the best closed model and the best available open model can move.

If a product or engineering team has built deeply around a single closed AI provider — hard-coded prompts, tooling, and pricing assumptions with no realistic path to switching — moments like this are a useful prompt to check that exit plan. Not because you should necessarily switch, but because the assumption that closed frontier models will always be meaningfully ahead is getting harder to take for granted with every generation.

Frequently Asked Questions

What is Kimi K3?
Kimi K3 is Moonshot AI’s flagship large language model, released July 16, 2026. It’s a 2.8-trillion-parameter Mixture-of-Experts model, activating 16 of 896 experts per token, with a 1-million-token context window and native vision support.

Is Kimi K3 open-source?
It’s marketed as open-weight, but as of its launch, full model weights had not yet been released. Moonshot AI promised the weight release by July 27, 2026. Until then, access is limited to the hosted API and Kimi’s own apps.

How does Kimi K3 compare to Claude Fable 5 and GPT-5.6 Sol?
Moonshot’s own published materials acknowledge K3 trails both Claude Fable 5 and GPT-5.6 Sol on general intelligence benchmarks. However, K3 ranked #1 on Arena’s crowd-judged Frontend Code leaderboard, ahead of Claude Fable 5 in that specific blind-comparison coding test.

How much does Kimi K3 cost to use?
Roughly $0.30 per million tokens for cache-hit input, up to $3 per million tokens for cache-miss input, and $15 per million tokens for output — notably cheaper than most closed frontier AI models.

How can developers try Kimi K3 right now?
Through Moonshot’s API using the model ID kimi-k3, which follows an OpenAI-compatible format, or through the Kimi consumer app at kimi.com. Self-hosting will only become possible once the open weights are released.

Share

The post A Free, Open AI Model Just Beat the Closed Frontier at Its Own Game appeared first on Blog.

ByteDance’s Doubao, Alibaba’s Qwen, and Tencent’s Yuanbao all currently let users build a custom AI persona — give it a name, shape its personality, and let it remember previous conversations. That feature is being switched off across all three platforms on the same day.

For anyone building AI products — not just in China, but anywhere — this is worth understanding closely. It is the clearest signal yet that “AI designed to feel human” is becoming its own regulated category, with real compliance costs attached.

What’s Actually Shutting Down

The feature in question is often called an AI companion or AI persona: a chatbot that isn’t just answering questions, but is designed to be talked to like a relationship. Users name it, describe its personality, and the AI carries memory of past conversations to make interactions feel continuous and personal.

Doubao, Qwen, and Yuanbao each built consumer-facing versions of this. It’s been a major driver of daily engagement for all three apps.

Starting July 15, that capability disappears. The rollout isn’t identical across platforms:

• Doubao users get a read-only export window through October 15, 2026, to save their chat history before it’s permanently removed.
• Qwen users have no announced export window. Alibaba has not published a data retention plan for existing companion conversations.
• Tencent’s Yuanbao is affected the same way, though Tencent has said less publicly about transition details.

For a feature used daily by a large share of each platform’s user base, this is a significant product change to make on a single fixed date.

The Law Behind the Shutdown

The trigger is a regulation from China’s Cyberspace Administration: the Interim Measures for the Administration of AI Anthropomorphic Interaction Services. It was issued in April 2026 and takes effect July 15.

Stripped of the legal language, the measure regulates any AI system designed to be human-like enough that users form real emotional attachments to it. That is a meaningfully different target than most AI regulation so far, which has focused on accuracy, bias, or data privacy. This law is about emotional design itself.

The requirements include:

• Anti-addiction mechanisms. Products must include break reminders and usage safeguards, similar to rules already applied to gaming apps in China.
• Protection for minors. AI companions cannot be positioned or designed to replace real relationships for users under 18.
• A ban on training on private conversations. Companies can no longer use users’ personal chat data, including companion conversations, to train models without explicit separate consent.
• Limits on manipulative design. Patterns that are built to increase emotional dependence — artificial urgency, guilt-based re-engagement prompts, and similar dark patterns — are restricted.

According to reporting from Bloomberg and TechTimes in early July 2026, both ByteDance and Alibaba concluded that patching their existing companion systems to meet these requirements in time wasn’t feasible. Rebuilding from the ground up, on a compliant architecture, was judged the faster path — even though it means switching the current feature off entirely first.

The Detail Most Coverage Is Missing

Most headlines have framed this as China “banning AI companions.” That framing misses the more interesting part.

China isn’t banning AI companionship as a category. It’s licensing it — setting a compliance bar high enough that only well-resourced companies can clear it. ByteDance is already reportedly redirecting users toward a separate, standalone companion app (reported under the name Maoxiang) built specifically to meet the new requirements from the ground up.

That distinction matters for market structure. Building anti-addiction systems, minor-protection logic, consent-based training pipelines, and manipulative-design audits is expensive engineering and legal work. ByteDance and Alibaba can absorb that cost and treat it as a rebuild. Smaller AI companion startups operating in China generally cannot.

The practical effect, according to analysis from FourWeekMBA, is consolidation. A law framed around user protection is also, functionally, a barrier to entry that favors the largest platforms in the market.

This Isn’t Happening in Isolation

China’s measure is the broadest version of a trend that’s already underway elsewhere.

In the United States, California’s SB 243 — effective January 1, 2026 — regulates companion AI chatbots specifically where minors are involved. Washington State’s HB 2225, effective January 1, 2027, goes further by banning manipulative engagement tactics designed to create emotional dependence, regardless of the user’s age.

China’s rule is broader than either of these because it applies to all users, not only minors. But the direction across all three jurisdictions is consistent: regulators are starting to treat emotionally engaging AI as a distinct governance category, separate from general AI safety or data privacy rules.

Regulators worldwide are converging on the idea that AI designed to feel human needs its own rulebook — not just an extension of existing data privacy law.

Why This Matters If You’re Building AI Products

Even if your product isn’t an AI companion app, the underlying questions this regulation raises are becoming standard product decisions for anyone building with conversational AI:

• Should this AI remember the user across sessions? Persistent memory is powerful for usability, but it’s exactly the kind of feature regulators are now scrutinizing.
• How much personality should this AI have? Products designed to feel warm and personal are more engaging — and increasingly, more regulated.
• What happens to user conversation data? Using chat logs to fine-tune or improve a model is common practice. It’s also now a specific compliance question in at least three jurisdictions.
• Are your engagement mechanics manipulative by design, even unintentionally? Notification timing, guilt-based prompts, and streak mechanics that were once purely growth tactics are now legal risk surfaces.

These aren’t hypothetical concerns for a future product cycle. They’re live design decisions for anything shipping conversational AI features today, particularly in fintech, healthcare, education, and consumer apps — sectors where emotionally resonant AI is often exactly the point.

How We Think About This at Onclick Innovations

We build AI features for clients across fintech, healthcare, e-commerce, and SaaS, and this shift changes how we approach a feature request before writing a line of code. Memory, personalization, and personality used to be almost purely UX decisions. They’re now compliance decisions with real legal weight behind them, and that changes how early in a project they need to be addressed.

Our approach on any AI feature with a personal or emotional dimension now includes mapping it against the relevant regulatory landscape for the client’s markets before development starts — not as a late-stage legal review, but as part of the initial architecture conversation.

Frequently Asked Questions

What is China’s new AI companion law?
It’s the Interim Measures for the Administration of AI Anthropomorphic Interaction Services, issued by China’s Cyberspace Administration in April 2026 and effective July 15, 2026. It regulates AI systems designed to be human-like enough to create emotional attachment, requiring anti-addiction safeguards, minor protections, restrictions on training with private conversation data, and limits on manipulative design.

Which companies are affected?
ByteDance (Doubao), Alibaba (Qwen), and Tencent (Yuanbao) are all shutting down their custom AI persona features as a direct result of the new rules taking effect.

Will users lose their AI companion chat history?
Doubao users have a read-only export window through October 15, 2026. Alibaba has not published a similar plan for Qwen users as of early July 2026.

Is China banning AI companion apps entirely?
No. The measures regulate rather than prohibit the category. ByteDance is reportedly building a separate, compliant standalone companion app, suggesting the category continues under stricter rules rather than disappearing.

Does this affect AI regulation outside China?
Not directly, but it reflects a broader global pattern. California’s SB 243 and Washington State’s HB 2225 regulate similar territory in the U.S., focused on companion AI and manipulative engagement design.

Share

The post China Just Banned AI Companions As You Know Them — Here’s What Every AI Product Team Should Learn From It appeared first on Blog.