The week in brief
Behind it, Promptfoo (+266) and Opik (+226) followed.
Four new tools joined the directory this week: GDPval, MMLU-Pro, mlprints and Fingerprinting Leaderboard.
On Hacker News, “90% of US working hours are spent on tasks not covered well by AI benchmarks” drew the most attention in this space (628 points, 575 comments). Among listed tools, Artificial Analysis's site was linked from 88 Hacker News stories over the past 90 days, more than any other tool here.
In research, “MMEnterprise-Bench: Evaluating Enterprise Agents over Evolving Multimodal Records” collected the most upvotes on Hugging Face Daily Papers this week (19). Among new GitHub repositories created in the past 7 days, liulx25xx/DoubleTakeBench leads with 11 stars.
Biggest GitHub star gainers
GitHub star counts for the 37 open source tools we track, compared between Oct 13, 2026 and Oct 21, 2026 (186 hours apart).
| Tool | Stars before | Stars after | Gain | Change |
|---|---|---|---|---|
| Langfuse | 35,767 | 36,129 | +362 | 1.01% |
| Promptfoo | 25,963 | 26,229 | +266 | 1.02% |
| Opik | 22,565 | 22,791 | +226 | 1.00% |
| DeepEval | 18,809 | 19,008 | +199 | 1.06% |
| Aider Leaderboards | 49,524 | 49,680 | +156 | 0.31% |
| Arize Phoenix | 11,819 | 11,934 | +115 | 0.97% |
| Ragas | 16,009 | 16,099 | +90 | 0.56% |
| garak | 9,553 | 9,634 | +81 | 0.85% |
New listings this week
| Tool | Category | Listed |
|---|---|---|
| GDPval | Benchmarks | Oct 17, 2026 |
| MMLU-Pro | Benchmarks | Oct 15, 2026 |
| mlprints | LLM Evals | Oct 15, 2026 |
| Fingerprinting Leaderboard | Leaderboards | Oct 15, 2026 |
Most discussed on Hacker News
The highest scoring Hacker News stories in this space during the week.
| Story | Points | Comments | Discussion |
|---|---|---|---|
| 90% of US working hours are spent on tasks not covered well by AI benchmarks | 628 | 575 | thread |
| Ontology-Driven Observability: Building an End to End Knowledge Graph at Netflix | 128 | 30 | thread |
| Benchmarking Bot Detection Systems Against Modern AI Agents in 2026 | 46 | 7 | thread |
| LabRobotBench v0.1: Benchmarking wet lab robotics | 13 | 1 | thread |
Listed tools with the most Hacker News links
Hacker News stories linking to each tool's own website over the past 90 days.
Notable new papers
The most upvoted papers on Hugging Face Daily Papers and the newest relevant arXiv preprints from this week.
- MMEnterprise-Bench: Evaluating Enterprise Agents over Evolving Multimodal Records
- EvaOmni: An Evidence-Grounded Benchmark for Audio-Visual Reasoning
- DecisionPivot: A Diagnostic Benchmark for Decision Models in LLM Systems
- EdgeKernelBench: Unified LLM Driven Kernel Generation for Edge Backends
- Beyond Compilation: Evaluating Faithful Natural-Language-to-Lean Statement Formalization
- MindView: Benchmarking Generative Multi View Spatial Intelligence
- Benchmarking Arabic Dialects for Math Problems
- What Do RTL Repair Benchmarks Actually Measure?
- DecisionPivot: A Diagnostic Benchmark for Decision Models in LLM Systems
- OptionWorld: A Benchmark for Diagnosing Failures of LLMs as Long Term Personal Assistants
New GitHub repos in LLM evaluation
Repositories created in the 7 days before Oct 21, 2026, ranked by stars.
| Repository | Stars | Language | Description |
|---|---|---|---|
| liulx25xx/DoubleTakeBench | 11 | Python | DoubleTakeBench: 525 evidence based judgment questions from coding agent trajectories for evaluating System 1 models and LLMs. |
| kling01/human-value-allocation | 9 | Python | Evidence linked analysis of human contributions in human and AI conversations. Research preview. |
| alphaparkinc/genpark-entropy-hallucination-detector-skill | 7 | Python | Token entropy and n-gram perplexity analyzer auditing LLM outputs for degenerate repetition loops and vocabulary collapse |
| Alpha-Park/genpark-entropy-hallucination-detector-skill | 7 | Python | Token entropy and n-gram perplexity analyzer auditing LLM outputs for degenerate repetition loops and vocabulary collapse |
| SummitAnthony/BanglaBBQ-benchmark | 4 | Python | BanglaBBQ: a bias benchmark for question answering in the Bangladeshi context (English and Bangla) |
| shashikantdev3/decision-models-lab | 4 | TypeScript | Learn decision models by building with them. TypeSafe Jev and Microsoft Decision-1 route support tickets in under a second for about $0.02 per 1,000 calls; Claude handles the hard cases. Interactive s |
| kling01/RubricLoop | 3 | Python | Rubric driven evaluation for AI agents. Executable criteria, code and model judges, and inspectable trace reports. |
| Cooops/kami-drills | 2 | Python | Offline decision drills for agents that play Kamigotchi |
How we compile this report
Every figure comes from a public source: star counts from the GitHub API, story scores and links from Hacker News (via Algolia), papers from arXiv and Hugging Face Daily Papers, and new repositories from GitHub search. Hacker News link counts include stories whose link points to the tool's own domain. Listing dates are the day a tool joined TrueIQ. Numbers are shown as recorded; if a figure is unavailable, it is left out rather than estimated.