Sign in
← Back to search
github Actively maintained

FareedKhan-dev/kimi-k3-in-c

A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.

1 awesome list

Quick read

Stars
7,128
Forks
1,159
Open issues
8
Commits
64

Activity and growth

Latest capture 2026-09-05 10:53

Stars · last 7 days
No history
Commits · last 7 days
No history
Stars since tracking
0
Stored snapshots
1

Classification

Topics

#avx2 #c99 #cpu-inference #deep-learning #from-scratch #inference-engine #kimi-k3 #linear-attention #llm #llm-inference #machine-learning #memory-efficient #mixture-of-experts #moe #mxfp4 #quantization #simd #systems-programming #transformer #zero-dependencies

Metadata

Language
C
License
Apache-2.0
Default branch
main
Created
2026-08-01
First commit
2026-08-01
Last pushed
2026-08-26
GitHub updated
2026-09-05
Last synced
2026-09-05 10:53
Stack scanned
2026-09-05 10:53
Archived
No
Sign in to like FareedKhan-dev/kimi-k3-in-c
README Open website View on GitHub

AI development signals

0 paths

Agent instructions and tool configuration found in this repository.

No config files detected.

Growth history

Tracked growth

1 observed capture since 2026-09-05. Observed captures are shown by default.

Stars from first capture 0

Chart data

Observed captures only

Time horizon

All tracked data

Custom date range

Stars history

Observed snapshots

Commits history

Observed snapshots

Similar repositories

Nearest indexed repositories by embedding similarity.

raullenchai/Rapid-MLX

The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separation, cloud routing. Drop-in OpenAI replacement. Works with Claude Code, Cursor, Aider.

3,700 stars
Python 1 awesome list

Scottcjn/legend-of-elya-n64

Legend of Elya — N64 game with a real 6.36M-parameter ternary transformer on the VR4300 MIPS III CPU. Zelda-style dungeon, AI NPCs, byte-level inference at 1.23 tok/s scalar / 2.19 tok/s on the RSP overlay (measured under ares, never on silicon). Built with libdragon.

128 stars
C 2 awesome lists

Blaizzy/mlx-vlm

MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) on your Mac using MLX.

5,467 stars
Python 1 awesome list

Mibayy/token-savior

MCP server that gets Claude to 97.9% (188/192) on a real coding benchmark at -80% active tokens and -83% wall time, vs 78.3% plain. Structural code navigation + persistent memory engine. Works with every MCP client.

1,140 stars
Python 2 awesome lists

ikamensh/kodo

Orchestrator for AI coding (claude code, cursor, codex, gemini)

130 stars
Python 1 awesome list