Kimi-Dev-72B
Parameters
72B
Context Length
131K
Modality
Text
Architecture
Dense
KV Cache Rate
320 KB/tok
License
MIT License
Release Date
16 Jun 2025
Knowledge Cutoff
-
API Pricing (per 1M)
Input: $0.29 · Output: $1.15
Hardware Sizing
Precision
Context
Workload Memory: 38.53 GB
INT4 · 8K context
Base Weights: 33.53 GB
KV Cache: 2.50 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
1x NVIDIA A40 48GB
38.5 / 43.2 GB usable
2x RTX 4090 24GB
38.5 / 43.2 GB usable · TP 2
Apple M4 Max 128GB
38.5 / 96.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Architecture Diagram
Evaluation Benchmarks
No evaluation benchmarks for Kimi-Dev-72B available.
Rankings
Overall Rank
-
Coding Rank
-
About Kimi-Dev-72B
Kimi-Dev-72B is an open-weights software engineering model by Moonshot AI built on Qwen2.5-72B and fine-tuned for autonomous issue resolution. Utilizing a dual BugFixer/TestWriter architecture, it specializes in codebase localization and verified code patches.
Technical Specifications
Attention
Attention Structure
Multi-Head Attention
Attention Heads
64
Key-Value Heads
8
Position Embedding
Absolute Position Embedding
RoPE Theta
1,000,000
Sliding Window Attention
No
Sliding Window Size
131,072
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Hidden Dimension Size
8,192
Number of Layers
80
FFN Intermediate Size (Dense)
29,568
Tokenizer
Vocabulary Size
152,064
About Kimi
Moonshot AI's Kimi model family, exemplified by Kimi K2, employs a Mixture-of-Experts architecture with one trillion total parameters. Designed for natural language generation and agentic capabilities, it features a 128K token context window. The models are open-weight and optimized with the Muon optimizer for stable training.
Other Kimi Models
- No related models available