Mindigenous
/

MINDI-1.5-Vision-Coder

code-generation

Model card Files Files and versions

MINDI-1.5-Vision-Coder / src

139 kB

Ctrl+K

Ctrl+K

2 contributors

History: 6 commits

Faaz

Fix Phase 2: fusion layer processes text-only via learnable residual gate for gradient flow

4e9835e about 1 month ago

agents
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago
evaluation
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago
inference
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago
model
Fix Phase 2: fusion layer processes text-only via learnable residual gate for gradient flow about 1 month ago
sandbox
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago
search
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago
tokenizer
Day 1 Complete: Tokenizer setup — Qwen2.5-Coder-7B base + 22 MINDI special tokens (vocab 151,685), wrapper class, full format test about 1 month ago
training
Fix: register LLM as nn.Module submodule so optimizer finds LoRA params about 1 month ago
utils
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago
__init__.py

219 Bytes
feat: initial project scaffold for MINDI 1.5 Vision-Coder about 2 months ago