Weekly benchmark review: 2026-07-06

[github-actions]

Weekly benchmark review (2026-07-06)

Automated check from scripts/weekly-benchmarks-check.mjs. Triage and either:

• Update data/benchmarks.json if a new flagship model dropped this week, then close this issue, OR
• Comment noop and close if nothing actionable surfaced.

Current state of data/benchmarks.json

• lastUpdated: 2026-06-11 (25 days ago)

• Models tracked: 20

• Benchmarks tracked: 5

• Models released within last 60 days: 1

  • 2026-06 | Anthropic | Claude Fable 5

Model-release-flavored news, last 7 days

Matched 14 articles (keyword scan; not all will be real releases).

Date	Source	Title
2026-07-06	Hacker News AI	What Claude Code Orchestrator Loops Look Like in Practice
2026-07-06	ZDNet AI	I tested the new Claude Desktop on Linux - here's how it compares to rival apps
2026-07-06	ZDNet AI	Claude Fable 5 is back, but I'm sticking with Opus 4.8 for daily work: 5 reasons why
2026-07-03	The Verge AI	Anthropic wants to develop its own drugs
2026-07-03	The Verge AI	A behind-the-scenes look at Midjourney’s medical scanner leaves many questions unanswered
2026-07-01	Google AI Blog	The latest AI news we announced in June 2026
2026-07-01	MIT Technology Review	LLMs are stuck in a groupthink groove. This startup is trying to get them out.
2026-07-01	The Verge AI	Google built a great smart speaker, but Gemini isn’t ready for it
2026-07-01	WIRED AI	Claude Helped a Hacker Find a Way to Issue Tickets to Almost Every US Music Festival
2026-07-01	The Verge AI	Anthropic’s long-sidelined Fable 5 is greenlit to return
2026-06-30	NVIDIA AI Blog	NVIDIA BioNeMo Agent Toolkit Brings Accelerated AI to Life Sciences Researchers in Claude Science
2026-06-30	Google AI Blog	Unlocking Britain’s next era of productivity: Building a nation of AI trailblazers
2026-06-29	WIRED AI	Meta Contractors Posed as Teens to Prompt Rival Chatbots About Suicide, Sex, and Drugs
2026-06-29	NVIDIA AI Blog	Claude Meets Blackwell Ultra: Anthropic’s Models Now Run on NVIDIA GB300 in Azure

Sources: The Verge AI (4), ZDNet AI (2), Google AI Blog (2), WIRED AI (2), NVIDIA AI Blog (2), Hacker News AI (1), MIT Technology Review (1)

HF Open LLM Leaderboard top 10

Captured: 2026-07-06

Rank	Model
1	MaziyarPanahi_calme-3.2-instruct-78b_bfloat16 (avg 52.08 · 78B)
2	MaziyarPanahi_calme-3.1-instruct-78b_bfloat16 (avg 51.29 · 78B)
3	dfurman_CalmeRys-78B-Orpo-v0.1_bfloat16 (avg 51.23 · 78B)
4	MaziyarPanahi_calme-2.4-rys-78b_bfloat16 (avg 50.77 · 78B)
5	huihui-ai_Qwen2.5-72B-Instruct-abliterated_bfloat16 (avg 48.11 · 73B)
6	Qwen_Qwen2.5-72B-Instruct_bfloat16 (avg 47.98 · 73B)
7	MaziyarPanahi_calme-2.1-qwen2.5-72b_bfloat16 (avg 47.86 · 73B)
8	newsbang_Homer-v1.0-Qwen2.5-72B_bfloat16 (avg 47.46 · 73B)
9	ehristoforu_qwen2.5-test-32b-it_bfloat16 (avg 47.37 · 33B)
10	Saxo_Linkbricks-Horizon-AI-Avengers-V1-32B_bfloat16 (avg 47.34 · 33B)

---

What "needs update" usually means

1. A flagship from Anthropic / OpenAI / Google / Meta / Mistral / DeepSeek / xAI launched this week → add a row to data/benchmarks.json.
2. A tracked model has materially-shifted benchmark scores (re-running, methodology change) → update the row.
3. A new benchmark itself (e.g. a successor to MMLU-Pro) is becoming canonical → add it.

What it usually does NOT mean

• Research papers about benchmarks (those land on /research, not /benchmarks).
• HN opinion threads about a model.
• Pricing-only changes (those go in data/pricing.json).

Bump lastUpdated in data/benchmarks.json whenever you change anything else in the file.