--- title: "LLM Fine-Tuning" description: "Fine-tune models on your codebase, conventions, and domain, so outputs match how your teams actually build." url: https://www.callstack.com/services/llm-fine-tuning updated: 2026-11-06T10:08:22.116Z --- # LLM Fine-Tuning Fine-tune models on your codebase, conventions, and domain, so outputs match how your teams actually build. _[Agentic Delivery](/agentic-delivery) › [Custom & Private Models](/capabilities/custom-private-models)_ **A model that writes code like your team** Generic models write generic code. We fine-tune models on your codebase, conventions, and domain - the same capability behind Apex, our specialized React Native model - so outputs match how your teams actually build. [Plan my model strategy](#contact) · [How it works](#what-you-get) ## Custom & Private Models - 30/31 Next.js tasks as pixel canary · 85% Lower inference cost vs Opus 5.5 - Industry runs on our tooling · 50M+ OSS downloads / month - 100+ Enterprise projects delivered - React Foundation Founding Member · 300+ commits to React Native · Contributing since 2016 - ISO 27001 certified - 32 Models tested against React Native ## What you get: Specialization that pays. - **Training data pipeline** Data curated from your repos, reviews, and docs, with licensing and privacy checks. - **Task-focused tuning** LoRA and full fine-tunes aimed at the workflows that matter, not vanity benchmarks. - **Rigorous evaluation** Improvements proven on held-out, real-task evals before rollout. - **Deployment flexibility** Tuned models served in your cloud, on-premise, or on-device. - **Cost-efficiency gains** Smaller specialized models beating larger generic ones on your tasks, at a fraction of the cost. - **Refresh process** A re-tuning cadence as your codebase and conventions evolve. **Stop prompting around the gap.** A model specialized to your stack, proven on your tasks. [Plan my model strategy](#contact) ## Scope → Build → Scale: Deploy models where the workload fits. ### Scope (1 WEEK) We select a product workflow, define where agents act, and agree the controls and success measures. - Workflow and owners defined - Baseline and controls agreed - Token budgets confirmed ### Build (1-12 MONTHS) We connect context, agents, and verification to your tools, then run the workflow on your product backlog. - Running on your product backlog - Speed and quality measured - Business impact validated ### Scale (AS NEEDED) We document the workflow, transfer ownership, and extend it across teams and products based on results. - Your team owns the workflow - Expand based on results - Full knowledge handover **Stop prompting around the gap.** A model specialized to your stack, proven on your tasks. [Plan my model strategy](#contact) > Their engineers proved to be a crucial part of our mobile app relaunch project. > > — Brian Aznar, Engineering Director / Major League Soccer We’ll connect you with the right expert within 24 hours. ## Let’s find the right path for your product. Still have questions? Whether you’re planning new work or unblocking an existing product, we’ll help you choose the right path forward. **When is fine-tuning worth it versus prompting or retrieval?** When prompting plateaus: persistent style mismatches, domain language the model lacks, or cost pressure at volume. We check the cheaper levers first and say so if they suffice. **How much data do we need?** Less than most expect. Thousands of quality examples move behavior; we build the dataset from your repos and review history with curation, not bulk export. **Which base models do you tune?** Open-weight models chosen by evaluation on your tasks - the same discipline we used building Apex, our React Native coding model. **Who owns the resulting model?** You do: weights, dataset, and evaluation harness. No per-seat licensing on your own model. **How long does a fine-tuning cycle take?** Weeks end to end: dataset construction, tuning runs, evaluation, and deployment. Refreshes are faster once the pipeline exists. **How do we know it is actually better?** Held-out evaluations on your real tasks, compared against the base model and hosted alternatives. If the tuned model does not win, we tell you. [Book consultation](#contact) ## Open source - [React Native AI](https://github.com/callstackincubator/ai): On-device LLM execution in React Native with Vercel AI SDK compatibility - [React Native Evals](https://github.com/callstackincubator/evals): A benchmark suite for evaluating how coding models solve real React Native tasks - [Skill Gym](https://incubator.callstack.com/skillgym/): Prove your agent skills work before you ship them Insights ## Worth your time, by engineers. [Explore Insights](https://www.callstack.com/insights) - [Conversational Banking: From Customer Questions to Authorized Actions](https://www.callstack.com/blog/conversational-banking-from-customer-questions-to-authorized-actions) (Nov 2 · Article): How does a banking assistant turn a customer’s question into an authorized action? Follow an unfamiliar payment from transaction lookup to card freeze and specialist handoff, exploring how models, banking services, and human teams work together and what the bank needs to control at each step. - [Apex Is Now Generally Available: Agentic Coding Model for Mobile and Web, Specialized in React](https://www.callstack.com/blog/apex-agetic-coding-for-mobile-and-web-spcialied-in-react) (Oct 1 · Article): Apex is now publicly available. Built by Callstack, it brings React Native expertise to feature development, code reviews, debugging, and testing, with strong performance on Next.js tasks too. See the benchmark results, compare API pricing, explore self-hosting options, and get started with Apex in your own project. - [On-device AI After WWDC 2026: What's New?](https://www.callstack.com/blog/on-device-ai-after-wwdc-2026-whats-new) (Jun 8 · Article): Apple's 2026 AI updates move Apple Intelligence beyond a narrow text model. Foundation Models gets image input and server-backed calls, high-end devices get a stronger on-device model, and Core AI gives custom local models a native path. We're mapping these APIs into the React Native AI stack and working with early users on real app workflows. - [Optimizing Self-Hosted Gemma for Production Inference](https://www.callstack.com/blog/optimizing-self-hosted-gemma-for-production-inference) (Jun 5 · Article): We optimized self-hosted Gemma 4 31B for Apex by tuning vLLM server configuration, GPU memory use, KV cache capacity, context length, multimodal limits, tensor parallelism, and MTP speculative decoding. The article covers what changed, the performance gains we measured, the latency tradeoffs, and the production setup we recommend for faster Gemma serving. - [Introducing Apex: A Fast, Specialized Model for React Native](https://www.callstack.com/blog/introducing-apex-a-fast-specialized-model-for-react-native) (May 27 · Article): Today, we are introducing Apex, our first domain-specific coding model designed for React Native. Built on Gemma 4, Apex achieves frontier-level performance at a fraction of the cost of generalized models. It is optimized for the fast, agentic workflows our teams run daily, delivering up-to-date results with fewer tool calls. - [When to Use Apple Foundation Models on Mobile](https://www.callstack.com/blog/when-to-use-apple-foundation-models-on-mobile) (Apr 27 · Article): Apple Foundation Models are most useful on mobile when the task is small, fast, local, and not worth a cloud round trip. The real value is not replacing cloud models entirely, but using on-device AI where latency, privacy, and frequency matter most. - [RAG Is Dead. Long Live Context Engineering for LLM Systems](https://www.callstack.com/blog/rag-is-dead-long-live-context-engineering-for-llm-systems) (Mar 27 · Article): Not every LLM system needs a full RAG stack. This article shows when structured context injection can replace retrieval pipelines, reducing complexity, latency, and operational overhead. - [A Practical Guide to LLM Model Naming Conventions](https://www.callstack.com/blog/how-quantization-and-naming-conventions-shape-llm-deployment) (Mar 18 · Article): The same LLM can appear in many variants because different hardware environments require different numerical precision. This article breaks down quantization, explains model naming conventions, and shows how deployment constraints shape model formats. Explore how quantization impacts performance and infrastructure costs. - [Announcing React Native Evals](https://www.callstack.com/blog/announcing-react-native-evals) (Mar 5 · Article): React Native Evals is a new open-source benchmark from Callstack that evaluates how coding models implement real React Native tasks. The suite ships with 39 evals across animation, async state, and navigation categories, covering libraries like Reanimated, TanStack Query, Zustand, and React Navigation. - [Profiling MLC-LLM’s OpenCL Backend on Android: Performance Insights](https://www.callstack.com/blog/profiling-mlc-llms-opencl-backend-on-android-performance-insights) (Dec 16 · Article): MLC-LLM can run LLMs fast on-device, but we hit a nasty Android issue: the first inference froze the system UI for up to 50 seconds, then everything was fine. This post walks through how we added Perfetto-compatible traces in MLC-LLM and kernel profiling in TVM/OpenCL, and why switching from \_1 to \_0 model formats fixed it on Adreno. - [On-Device Text To Speech on Apple Devices with AI SDK](https://www.callstack.com/blog/on-device-text-to-speech-on-apple-devices-with-ai-sdk) (Aug 21 · Article): We’ve added on-device speech synthesis to our Apple provider for the AI SDK, completing the feature set alongside generation, embeddings, and transcription. This integration with AVSpeechSynthesizer brings fast, private, and offline text-to-speech to React Native, with support for multiple voices, languages, and even Personal Voice on iOS 17+. - [On-Device Speech Transcription in React Native with Apple SpeechAnalyzer](https://www.callstack.com/blog/on-device-speech-transcription-with-apple-speechanalyzer) (Aug 6 · Article): This post introduces speech transcription support in React Native AI, implemented using Apple’s SpeechAnalyzer and SpeechTranscriber APIs available in iOS 26. It explains how to use the AI SDK’s transcribe function to pass audio buffers, how modular analysis is configured through the SpeechAnalyzer class, and how system-level asset management ensures models are handled outside the app bundle. - [On-Device Text Embeddings in React Native With Apple NLP Framework](https://www.callstack.com/blog/on-device-ai-introducing-apple-embeddings-in-react-native) (Jul 30 · Article): Our React Native AI library now supports on-device text embeddings by integrating Apple's native Natural Language framework. This implementation leverages the NLContextualEmbedding model built directly into iOS. Because the model is treated as a shared operating system asset, it enables powerful semantic features without requiring developers to bundle large files or force users to perform any extra downloads. We expose this capability through our Vercel AI SDK provider, simplifying integration for React Native developers. - [Expanding On-Device Apple LLM Capabilities: Introducing Tool Calling](https://www.callstack.com/blog/expanding-on-device-apple-llm-capabilities-introducing-tool-calling) (Jul 25 · Article): We’ve just added tool calling to our Apple provider for the AI SDK, unlocking the ability to run custom functions directly within Apple’s on-device language model. This means your React Native apps can now generate dynamic, interactive content, like fetching live data or executing app-specific logic, without ever leaving the device. - [On-Device Apple LLM Support Comes to React Native](https://www.callstack.com/blog/on-device-apple-llm-support-comes-to-react-native) (Jul 14 · Article): Apple’s on-device LLM is now available in React Native, enabling developers to build private, fast, and offline AI features powered by Foundation Models. This preview release supports text generation, streaming, structured outputs, and first-class support for Vercel AI SDK. A stable version is planned for release alongside iOS 26 later this fall. - [Want to Run LLMs on Your Device? Meet MLC](https://www.callstack.com/blog/want-to-run-llms-on-your-device-meet-mlc) (May 28 · Article): Running LLMs locally on mobile isn’t a fantasy-it’s possible with MLC. This post explains how MLC compiles both model and runtime to create optimized, cross-platform packages for local inference. Discover what’s required, how it works, and why MLC stands out among edge ML frameworks. - [Meet react-native-ai: LLMs Running on Mobile, for Real](https://www.callstack.com/blog/meet-react-native-ai-llms-running-on-mobile-for-real) (May 15 · Article): Running AI models locally in React Native used to sound impossible, but not anymore. This post introduces react-native-ai, a new library powered by MLC and integrated with Vercel AI SDK, enabling on-device LLM inference with a web-like API and native performance. - [Local LLMs on Mobile Are a Gimmick-For Now](https://www.callstack.com/blog/local-llms-on-mobile-are-a-gimmick) (Mar 11 · Article): Running large language models directly on mobile devices is currently limited by model size, format compatibility, and performance trade-offs. However, tools like MLC are emerging to simplify the process and pave the way for a future where everyone has their own AI assistant in their pocket. - [Creating a Video Transcription App: Lessons Learned](https://www.callstack.com/blog/creating-a-video-transcription-app-lessons-learned) (Dec 1 · Article): Callstack shares lessons learned from building a video transcription app using OpenAI's Whisper for speech recognition. The guide covers extracting audio, generating transcripts, speaker diarization, and testing different Whisper model sizes to optimize for speed and accuracy. - [The Hard Way vs. The React Native AI SDK Way: Stop Writing Custom Modules for Every Model](https://www.callstack.com/tutorials/the-hard-way-vs-the-react-native-ai-sdk-way-stop-writing-custom-modules-for-every-model) (Mar 19 · Tutorial): Learn how React Native AI uses a unified provider model to switch between cloud and on-device LLMs without rewriting application logic, enabling reliable AI experiences online and offline. - [Build Smarter Apps: Tool Calling & AI Orchestration Explained](https://www.callstack.com/tutorials/build-smarter-apps-tool-calling-ai-orchestration-explained) (Mar 12 · Tutorial): Learn how on-device models in React Native can call tools, interact with external APIs, and return structured outputs you can use directly in application logic. - [MLC LLM + React Native: On-Device AI Without the Pain](https://www.callstack.com/tutorials/mlc-llm-react-native-on-device-ai-without-the-pain) (Mar 5 · Tutorial): Learn how to run third-party on-device LLMs in React Native using MLC LLM. Choose your models, run them efficiently across platforms, and keep a unified JavaScript API with React Native AI. - [Intelligent Fallbacks with Apple Intelligence in React Native](https://www.callstack.com/tutorials/intelligent-fallbacks-with-apple-intelligence-in-react-native) (Feb 24 · Tutorial): Learn how to use Apple’s built-in Foundation Models in React Native apps, reduce memory usage, and enable fast, fully on-device AI with React Native AI Apple. - [What Is the React Native AI SDK? A Complete Intro & Quickstart](https://www.callstack.com/tutorials/what-is-the-react-native-ai-sdk-a-complete-intro-quickstart) (Feb 10 · Tutorial): Learn why on-device AI matters for React Native apps, how local LLMs behave offline, and what problems React Native AI is designed to solve from day one. - [Quantization Impact on LLM Quality: A Case Study of Apex](https://www.callstack.com/events/quantization-impact-on-llm-quality-a-case-study-of-apex-a-gemma-based-model) (Nov 4 · Talk): Artur Morys-Magiera and Lech Kalinowski share findings from 344 benchmark runs with Apex, our Gemma-based LLM, examining the trade-offs between model size, decode speed, and quality. - [Can Jev Make AI Agent Decisions Faster?](https://www.callstack.com/events/can-jev-make-ai-agent-decisions-faster) (Oct 1 · Live Stream): Explore Jev with Kewin Wereszczyński and Szymon Chmal: how it works, its approach to faster agent decisions, and possible local alternatives. - [No Framework, No Server: Making AI Agents Collaborate in One Terminal](https://www.callstack.com/events/no-framework-no-server-making-ai-agents-collaborate-in-one-terminal) (Sep 15 · Talk): Lech Kalinowski will show how Agents Commander lets multiple AI coding agents collaborate directly in one terminal without a central server or heavyweight framework. - [What Apex shows about React Native AI models](https://www.callstack.com/events/can-a-specialized-model-beat-frontier-models-at-react-native) (Aug 13 · Live Stream): See how Apex, Callstack's React Native model, compares with frontier models on navigation, animations, async state, and performance-heavy UI work. - [Ambient Generative AI: Deploying Latent Diffusion Models on Mobile NPUs](https://www.callstack.com/events/ambient-generative-ai-deploying-latent-diffusion-models-on-mobile-npus) (Apr 20 · Talk): Join Lech Kalinowski at AI Engineer Miami on April 20–21 for a talk on running offline latent diffusion models on mobile NPUs. - [React Native Evals: Measuring AI Code Quality in Practice](https://www.callstack.com/events/react-native-evals-making-ai-code-quality-measurable) (Mar 12 · Live Stream): React Native Evals introduces a data-driven way to measure how AI coding models perform on real React Native development tasks. - [LLM Inference On-Device in React Native: The Practical Aspects ](https://www.callstack.com/events/llm-inference-on-device-in-react-native-the-practical-aspects) (Jan 29 · Talk): A practical look at reliability, performance, libraries, and tradeoffs when running LLM inference locally in React Native apps. - [The Offline AI: On-Device LLMs in React Native With AI SDK](https://www.callstack.com/events/the-offline-ai-on-device-llms-in-react-native-with-ai-sdk) (Dec 5 · Talk): Learn how to run on-device LLMs in React Native using Vercel’s AI SDK from Michał Pierzchała's talk at DevAI by Data Science Summit. - [How to Run Any LLM On-Device With React Native](https://www.callstack.com/events/how-to-run-any-llm-on-device-with-react-native) (Nov 28 · Talk): In this talk, Szymon will show how to run LLMs directly inside React Native apps using an AI SDK that provides a powerful abstraction layer to simplify building AI applications. Join him as he explores react-native-ai, a library that enables local LLM execution. - [React Native AI: Bringing On-Device LLMs With AI SDK](https://www.callstack.com/events/react-native-ai-bringing-on-device-llms-with-ai-sdk) (Sep 2 · Talk): Need offline AI in your mobile app? This talk recording demonstrates how to use the AI SDK to create a custom provider for on-device LLMs in React Native, ensuring privacy and functionality anywhere. - [AI Meetup in Wroclaw with Callstack & Vercel](https://www.callstack.com/events/ai-meetup-in-wroclaw-first-edition) (Sep 2 · Meetup): On Sept 2 in Wrocław, join top speakers from around the world at the AI Meetup. Learn about AI SDK 5, on-device LLMs in React Native, TV-ready apps with Kiro.dev, and how AI supports developers beyond code. - [Live Dev Session: First Impressions of Apple’s On-Device AI](https://www.callstack.com/events/live-dev-session-react-universe-news-tools-code) (Jun 10 · Live Stream): Tune in to dev chat on Apple’s AI SDK, iOS 26 naming chaos, Siri’s limits, Liquid Glass UI, and on-device LLMs. - [Run LLMs Locally With React Native & MLC](https://www.callstack.com/events/run-llms-locally-with-react-native-mlc) (Jun 3 · Live Stream): Watch how MLC brings on-device AI to React Native. See react-native-ai in action and discover what’s next for local model execution. - [Bringing AI to React Native](https://www.callstack.com/events/bringing-ai-to-react-native) (Jul 25 · Talk): Leveraging capabilities of local devices to run LLMs via React Native and Vercel AI SDK.