krisman.dev | Projects

BlackwellQwen38

Qwen 3.8 27B at 1M context on one Blackwell GPU, with Claude Code, WebSearch, and hour-scale video working end to end. Six compose variants from 96 GB down to 24 GB cards, zero monkeypatches.

vLLMLiteLLMtransformersQwen 3.8Claude Code+2

View project

shippedrepo

litellm-claude-code-websearch

A LiteLLM plugin that makes Claude Code's WebSearch tool work against local models, running the search through Brave and answering in the native Anthropic block format.

PythonLiteLLMBrave Search APIClaude Code

View project

activerepo

gobig2

Pure-Go JBIG2 decoder. Within 1.3–1.6× of jbig2dec on every fixture, no cgo, with hard caps on every attacker-controlled allocation.

GoJBIG2ITU-T T.88PDFCGO-free

View project