Skip to content
Make in India OEM · INR-transparent · Pan-India onsite SLATalk to sales: +91 720 794 8743Sign in
Knowledge Base AI Architectures Fine-tuning

Fine-tuning

9 articles
Concept 28 Jul 2026

Continual Pre-Training: When Fine-Tuning Is Not Enough

Fine-tuning teaches behaviour; continual pre-training teaches a language or a domain's underlying distribution. It costs orders of magnitude more and risks catastrophic forgetting. This sets out when it is genuinely…

7 min readRead →
How-to 28 Jul 2026

Indian-Language Adaptation: Tokenizers, Data and GPU Planning

Sovereign Indian models from Sarvam, BharatGen and Gnani now cover 22 languages, and IndiaAI has commissioned over 36,000 GPUs heading toward 100,000. This covers tokenizer efficiency, data strategy and how…

7 min readRead →
Sizing guide 28 Jul 2026

Sizing the RL Fine-Tuning Loop: Rollouts, Verifiers and GPU Split

Reinforcement learning with verifiable rewards is now standard post-training, and it inverts the usual sizing assumption: most of the GPU budget goes to generating rollouts, not to gradient steps. That…

7 min readRead →
Explainer 28 Jul 2026

Fine-Tuning Mixture-of-Experts Models: Memory and Routing Realities

MoE models activate few parameters but must hold all experts in memory. A widely cited example needs roughly 94 GB at FP16 for inference alone, and fine-tuning adds optimiser state…

7 min readRead →
15 Jul 2026

Beyond SFT: Sizing DPO and RLVR Post-Training Infrastructure

The 2026 post-training recipe is SFT, then DPO, then RL with verifiable rewards. DPO doubles resident model copies; GRPO halved RL memory by dropping the critic, putting 7-32B reasoning training…

5 min readRead →
12 Jul 2026

GPU Cluster Networking for Training and Fine-Tuning

GPU cluster networking is the critical path for distributed training and fine-tuning: the interconnect fabric between GPUs determines whether your cluster scales linearly or stalls on collective communication. For workloads…

9 min readRead →
8 Jul 2026

Fine-Tuning GPU Server Sizing for Enterprise LLMs

Fine-tuning infrastructure should be sized around dataset shape, experiment cadence, checkpoint strategy, and governance controls before GPU count. RDP GPU Mart can help Indian teams compare DRACO GPU server options…

4 min readRead →
6 Jul 2026

Best On-Prem Setup for a Startup Training Small LLMs (<13B)

A startup working with sub-13-billion-parameter models needs far less hardware than the headlines suggest: a single GPU workstation with 48–141 GB of VRAM handles fine-tuning (QLoRA/LoRA) and inference for models…

4 min readRead →
6 Jul 2026

How Many GPUs Do You Need to Fine-Tune a 70B LLM On-Prem?

Fine-tuning a 70-billion-parameter LLM on-premises needs roughly one GPU for QLoRA, two GPUs for standard LoRA, and eight or more for full FP16 fine-tuning. The deciding variable is the tuning…

7 min readRead →

Need help in Fine-tuning?

Request a Quote