The week in brief
Behind it, Langfuse (+311) and Opik (+164) followed.
10 new tools joined the directory this week: SO-Arena, QuillBot, ZeroGPT, Vals AI and MKA (Multimodal Knowledge Agent) and 5 more.
On Hacker News, “Perplexity Decider v1.3: Training, eval, and use cases” drew the most attention in this space (147 points, 26 comments). Among listed tools, Artificial Analysis's site was linked from 90 Hacker News stories over the past 90 days, more than any other tool here.
In research, “SO-Arena: Benchmarking the Capability Spectrum and Reliability of System One Models” collected the most upvotes on Hugging Face Daily Papers this week (85). Among new GitHub repositories created in the past 7 days, jackson-cmd/risk-aware-trading-agents leads with 201 stars.
Biggest GitHub star gainers
GitHub star counts for the 42 open source tools we track, compared between Nov 9, 2026 and Nov 16, 2026 (175 hours apart).
| Tool | Stars before | Stars after | Gain | Change |
|---|---|---|---|---|
| Promptfoo | 26,981 | 27,322 | +341 | 1.26% |
| Langfuse | 37,185 | 37,496 | +311 | 0.84% |
| Opik | 23,278 | 23,442 | +164 | 0.70% |
| DeepEval | 19,381 | 19,512 | +131 | 0.68% |
| Aider Leaderboards | 50,030 | 50,143 | +113 | 0.23% |
| PyRIT | 4,773 | 4,880 | +107 | 2.24% |
| Arize Phoenix | 12,211 | 12,317 | +106 | 0.87% |
| garak | 9,962 | 10,051 | +89 | 0.89% |
New listings this week
| Tool | Category | Listed |
|---|---|---|
| SO-Arena | Benchmarks | Nov 14, 2026 |
| QuillBot | Integrity & Detection | Nov 14, 2026 |
| ZeroGPT | Integrity & Detection | Nov 13, 2026 |
| Vals AI | Leaderboards | Nov 13, 2026 |
| MKA (Multimodal Knowledge Agent) | Benchmarks | Nov 13, 2026 |
| Jev Decision Index | Leaderboards | Nov 11, 2026 |
| Harness Leaderboard | Leaderboards | Nov 10, 2026 |
| Copyleaks | Integrity & Detection | Nov 10, 2026 |
| Winston AI | Integrity & Detection | Nov 09, 2026 |
| Jev (TypeSafe AI) | LLM Evals | Nov 09, 2026 |
Most discussed on Hacker News
The highest scoring Hacker News stories in this space during the week.
| Story | Points | Comments | Discussion |
|---|---|---|---|
| Perplexity Decider v1.3: Training, eval, and use cases | 147 | 26 | thread |
| Next Level Benchmarking Overfitting: Thomas Neumann | 68 | 13 | thread |
| Eval Awareness vs. Eval Paranoia | 61 | 29 | thread |
| Ask HN: How do you evaluate the skills your AI agents use? | 25 | 8 | 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.
- SO-Arena: Benchmarking the Capability Spectrum and Reliability of System One Models
- VibePrevisBench: Evaluating Agentic Coding for 4D Scene Generation in Blender
- Actor, Judge, or Teacher? Understanding Decision Models in Long Horizon Agents
- TTB: Test time Baking for Efficient Rendering of Decoder only View Synthesis Models
- Do Underwater Image Quality Metrics Agree with Human Perception? A Benchmark with Human Choices, Simple Edits and Generative Drift
- WeightBench: Evaluating Language Models as Weight-Space Programmers
- MeReader: An Offline, Privacy Preserving, Progress-Aware AI Assistant for Narrative eBook Reading
- TradeGauntlet: Evaluating Compliance Controls for Autonomous Trading Agents
- Evaluating Feedback-Driven Black Box Processes via Stochastic Modeling and Simulation
- Do Verifier Judgments Survive Abstraction? Measuring Verification Transport for LLM Agent Evaluation
New GitHub repos in LLM evaluation
Repositories created in the 7 days before Nov 16, 2026, ranked by stars.
| Repository | Stars | Language | Description |
|---|---|---|---|
| jackson-cmd/risk-aware-trading-agents | 201 | Python | AI trading agents for US stocks with news backtesting that can't see the future: a public LLM trading agent leaderboard (rta-bench-1), frozen point in time 2020/2022/2026 markets (prices, VIX, news), |
| Sid00011/Flagwake | 6 | Python | A free, offline, recheckable test of how an AI defender behaves during a security incident, scored with business cost. |
| chessifyme/gmbench | 4 | Python | Benchmark of how well AI and LLM agents explain chess: positions, moves and plans, graded against expert answer keys |
| ranajahanzaib/finetuning-regression-tester | 4 | Python | Evaluate fine tuned language models for task gains and behavioral regressions with paired tests. |
| DawitLemmaBalcha/code-review-bench | 4 | Python | Review benchmark for model generated code patches: six tasks across Python, JavaScript, Go and Java with hidden tests, a fixed rubric and a Docker runner. |
| anoopcodehack/MedThread | 4 | TypeScript | MedThread , A clinical grade data synthesis platform that transforms heterogeneous, unstructured medical documents into a verifiable, dated longitudinal health timeline, with zero hallucinations, str |
| xzx34/RewardAudit | 3 | Python | Automated testing of rule based verifiers (checkers, reward functions) for LLM evaluation and RLVR. No model, no GPU. |
| aliraza2025/receipts | 3 | Python | Evidence first PM skills for Claude Code, Codex, and Cursor: make PRDs, metrics, experiments, AI evals, and launch copy show their receipts. |
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.