LogoCookLLM
  • Docs
  • Blog
  • Pricing
  • Roadmap
  • Commits
  • About Me
LogoCookLLM

Go deep. Build real things.

TwitterX (Twitter)Discord小红书Email
Product
  • Pricing
  • FAQ
Resources
  • Blog
  • Docs
  • Roadmap
  • Commits
Company
  • About Me
  • Contact Us
Legal
  • Cookie Policy
  • Privacy Policy
  • Terms of Service
Dang.aiStartup FameNewTool.siteFazierTwelve ToolsTurbo0
© 2026 CookLLM All Rights Reserved.

Roadmap

Explore our future plans and upcoming course modules

Overall progress14/29 completed

48%

Backlog
9
Modern Transformer tricks doclow
HarryHarry
KV Cache & text generation docmedium
HarryHarry
Evaluation metrics doclow
HarryHarry
RLHF: PPO training doc and codemedium
HarryHarry
DPO (Direct Preference Optimization) doc and codemedium
HarryHarry
GRPO (Group Relative Policy Optimization) doc and codemedium
HarryHarry
Build a mini vLLM inference enginehigh
HarryHarry
Multimodal model doc and codemedium
HarryHarry
LLM Agent framework doc and codemedium
HarryHarry
In Progress
6
Introduce modern LLM model architectures designshigh
HarryHarry
Normalization doc (BatchNorm → RMSNorm)medium
HarryHarry
Activation functions doc (ReLU → SwiGLU)medium
HarryHarry
Optimizer theory doc (Adam → Muon)medium
HarryHarry
Training techniques doc (LR schedule, mixed precision)medium
HarryHarry
SFT (Supervised Fine-Tuning) doc and codemedium
HarryHarry
Done
14
Introduce LLM tokenizerhigh
HarryHarry2026-01-17
Training tokenizer from scratchhigh
HarryHarry2026-01-17
Flash Attention doc and codehigh
HarryHarry2025-12-15
GPU programming doc and codehigh
HarryHarry2025-12-15
RoPE position encoding doc and codehigh
HarryHarry2026-02-10
build the cookllm-bento codebasehigh
HarryHarry2026-01-28
Training: tokenizer hands-on (RustBPE + tiktoken)high
HarryHarry2026-02-05
Training: BentoLM architecture design & implementationhigh
HarryHarry2026-02-14
Training: pretrain pipeline (data, loop, monitoring)high
HarryHarry2026-02-20
Training: pretrain inference & text generationmedium
HarryHarry2026-02-22
Introduce training parallelism modulehigh
HarryHarry2026-03-12
Transformer LM doc (overview, embedding & LM head)high
HarryHarry2026-04-08
Attention mechanism doc (+ attention sink)medium
HarryHarry2026-05-14
Coding harness doc & code (cookllm-harness)high
HarryHarry2026-06-06