AI/ML●●Solid
Breathe-Memory – Associative memory injection for LLMs (not RAG)
Graph-based context compression beats lossy summarization when tokens run out.
Big BrainNiche Gem
mvyshnyvetska
612mo ago
Back to browse
A 150M model that extracts verbatim evidence spans for RAG, no LLM call
by justacoolname·Jun 10, 2026·6 points·0 comments
AI Analysis
●●SolidNiche GemBig BrainSolve My Problem
150M model replaces LLM calls for evidence extraction with comparable F1 scores.
Strengths
- •150M params vs LLM calls means deterministic, cheap, local inference for production RAG.
- •Trained on financial tables, legal contracts, medical docs—not just Wikipedia QA like competitors.
- •8192 token ModernBERT context handles long passages without chunking overhead.
Weaknesses
- •Other extractors exist (Provence, Zilliz Semantic-Highlight)—category isn't empty.
- •Benchmark claims focus on ACL gold; real-world RAG performance less proven.
Category
Target Audience
ML engineers building RAG systems, teams needing cheaper evidence extraction than LLM calls
Similar To
Zilliz Semantic-Highlight · Provence · MultiSpanQA
Similar Projects
Developer Tools●●Solid
Argmin AI, system level LLM cost optimization for agents and RAG
LLM cost optimizer, but Anthropic's batch API and local quantization solve this cheaper.
Solve My ProblemBig Brain
konyrevdmitriy
203mo ago
AI/ML●●Solid
A tool to create and evaluate document processing pipelines for RAG
LLM-as-judge metrics beat guessing chunk sizes, but Ragas and LangSmith already exist.
Solve My ProblemSlick
martimchaves
202mo ago
AI/ML●●●Banger
I made a small helper for checking model-graded answers
Structurally verifies LLM judge reasoning instead of paying for a second model check.
Big BrainSolve My ProblemDark Horse
ML0037
204d ago
AI/ML●●Solid
Portable offline LLM knowledge system that runs in browser
Single-file RAG bundle runs entirely in browser without server setup.
Niche GemShip It
muthuishere
102mo ago
AI/ML●●Solid
Nexa-Gauge – LLM eval framework, now with self-hosted model support
Cache-aware LLM eval with self-hosted model support beats Ragas on flexibility.
Solve My ProblemSlick
Sardhendu
201mo ago