Skip to content
All tags

#benchmark

16 posts

How AI Agent Search Infrastructure Is Changing: Keenable, Independent Indexes, and NEEDLE

Keenable.ai positions itself as search infrastructure for AI agents: a 100B+ document index, Search/Fetch APIs, MCP/CLI entry points, 100K free monthly requests, and keyless public endpoints. It is worth tracking, but the 100B+ index, latency, and quality claims are still mostly company-provided; NEEDLE is open, but needs external reruns and human review.

How to Read 2026 Coding Benchmarks: SWE-bench, Terminal-Bench, DeepSWE, Aider Explained

The same model can score 20 points apart on different harnesses, 32% of SWE-bench Pro verifier judgments were found to be wrong, and DeepSWE's 113 tasks make most models score zero. This guide decodes six major coding benchmarks — what they test, which are easy to game, and which ones you should care about.

Why MoE Wins: The Architecture Behind Every 2026 Frontier Model

Nearly every frontier open-source model in 2026 is MoE: Ornith 35B activates only 3B to beat 31B dense models, MiniMax M3 uses 456B total but 45.9B active to hit SWE-bench Pro 59%, DeepSeek V4 runs 1.6T total with 49B active. This post explains why MoE dominates coding and agentic benchmarks using four case studies.

How to Read a Model's Report Card: Benchmarks, Arena Elo, and the Traps Behind the Numbers

Benchmark scores in model releases have three common traps: cherry-picking (only showing wins), contamination (test data leaking into training), and saturation (when everyone scores 90%+, the benchmark stops being useful). The most manipulation-resistant signal is Chatbot Arena's Elo ranking — real humans, blind voting, uncontrolled questions.

MiniMax: The Chat App Company That Built a Coding Model to Rival Frontier Labs

MiniMax started as a consumer chat app company, then M2.5 scored 80.2% on SWE-bench Verified at 1/10-1/20 the cost of Claude Opus; M3 (456B total / 45.9B active) became the first open-weight model to clear 59% on SWE-bench Pro, with 1M context powered by their novel Sparse Attention mechanism.

Ornith: The Open-Source Coding Dark Horse Built on Self-Improvement RL

DeepReinforce's Ornith 1.5 family, trained with self-improvement RL: the 397B flagship scores 86.0 on SWE-bench Verified, matching Claude Opus 4.8; the 35B-A3B activates only 3B parameters per token yet leads every coding benchmark in its class; the 9B runs on phones. MIT-licensed, fully open-source.

Learning Design from Mature Coding Agents (17): Startup Performance and Engineering Discipline — It Was Never the Language

A CLI tool pays its startup cost on every invocation, and performance optimization without a baseline means no regression protection. codex uses daemon reuse and skill snapshot caches; claude-code splits its entrypoint into dynamic imports plus a built-in startup profiler; opencode and omp each maintain lazy-loading discipline; pi does none of it and leans on Bun being fast. rivumi is Python — slow by birth — so it applies the full discipline: lazy imports, single-flight disk cache, background controller prewarming, and hyperfine paired benchmarks wired to a CI gate that fails on >10% regression.

tech guide

AI Model Evaluation Sources: How to Judge Whether a Model Is Actually Good

You cannot take model vendors' self-reported scores at face value. This guide covers the most important independent evaluation platforms, domain benchmarks, adoption indicators, and official sources in 2026: what each measures, how to read it, where it is biased, and which figures matter for different use cases.

CS224N Lecture 11: Why LLM Benchmarks Expire

Lecture 11 divides evaluation into what to test, how to measure it, and when the result stops being trustworthy. Benchmarks saturate or leak, prompts change scores, and an LLM judge remains a biased model.

CS336 Lecture 12: There Is No Single True LLM Score, Only Different Games

Lecture 12 moves from perplexity to exams, chat preferences, agents, reasoning, and safety. Every benchmark changes the capability definition, scaffold, judge, and contamination risk, so evaluation must first say whether it compares a method, model, or complete system.

Private-Corpus Retrieval Eval: Turning a Traditional Chinese Query Set into a Reproducible Benchmark

The repository has a 20-query Traditional Chinese/English golden dataset, but no document-level qrels, retrieval runs, raw latency data, or executable benchmark script. Reporting Recall@k, MRR, or nDCG as measured results would therefore be dishonest; this article defines the contract needed to run them reproducibly.

Web Extraction Quality Benchmark: Crawl4AI, Firecrawl, Jina Reader, and Readability

Extraction tools cannot be compared by HTTP 200s. The same 20 URLs must be scored for body text, headings, tables, code, links, metadata, noise, latency, and cost. This article publishes the corpus, adapter contract, and gates, but no winner without a same-version raw run across all four paths.

How to Evaluate Agent Search Quality: Building a Web Retrieval Benchmark

A web retrieval benchmark must evaluate complete tasks, not HTTP 200s: 30 fixed cases across five failure strata and three live channels, measuring answers, citations, freshness, latency, cost, and unnecessary escalation. This article delivers the harness and gates, but no fabricated ranking while the three live channels remain unconfigured.

ai deep-dive

Multimodal Models, First Half of 2026: Native Fusion vs. Bolted-On Vision, and Why Leaderboards Contradict Each Other

Pure image understanding has flattened out — four frontier models all clear 80% on MMMU-Pro within 3 points of each other. The real differentiation is video, long-document OCR, and realtime speech, each with a different leader. But the most useful lesson from assembling these rankings is that two credible sources named different Video-MME leaders more than 10 points apart — and that July and August each turned the field over again.

What's Going On Inside My Model? Where You Look First When It Regresses

Ask a model what a goose looks like to it and it draws a whole flock — because the labeled data tagged a flock as 'goose,' so it thinks the flock is the label. This lecture gives seven ways to open a CNN up, then says honestly: applied to transformers, even the frontier of this research only explains two layers.

Scanned PDF Benchmark: How Did 10 Parsers Handle Graduate Entrance Exams?

I tested 10 open-source PDF parsing tools on four scanned NTU graduate entrance exams. VLM-based tools—Firecrawl, MinerU 3.4, and Marker v2—overwhelmingly beat conventional OCR on formulas and code, but installation was the real barrier: MinerU's old package name creates dependency hell, Marker's first model download takes 10 minutes, and PaddleOCR needs a separate engine. In practice, use RapidOCR for screening and MinerU or Firecrawl for close inspection.