Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-09-05 · source-backed
The v4.2 update published September 4 adds AA-Briefcase, a private eval of agentic knowledge work on expert-built projects, and GDP.pdf, a document-reasoning test over 4,592 pages of tables, charts and footnotes. GPQA Diamond was dropped as saturated. Fable 5.1 leads overall, Astra gained about 85 Elo over its predecessor and dominates GDP.pdf at 33.2%. Doubling the held-out weighting is the signal: public benchmarks are now being treated as compromised (Artificial Analysis).
Each link below shares sources, entities, or timing with this story.
Artificial Analysis has Grok 4.6 at 61, one point under Fable 5 Max's 62, leading GDPval-AA v2 at 1753 (vs 1741 and 1728) and AA-Briefcase at 1577 (vs 1574 and 1502), beating Sol on 6 of 9 shared benchmarks. Then Terminal-Bench v3.0: 26% versus 34.6% for Sol and 34.1% for Fabl...
The Astra coverage went to price and context window. The number that changes how I'd deploy it went into a system card nobody read. Artificial Analysis measured GPT-6 Astra's hallucination rate on AA-Omniscience at 51% at max effort, against 92% for its predecessor. Accuracy w...
67 on coding against Fable 5.1's 70 in Claude Code. Astra does post a 2% hallucination rate against 9.4% for GPT-5.6 Sol, and 0% scope violations against 48%. Per-task cost runs the other way, $4.72 for Astra against $9.18 for Fable 5.1 at identical $10/$50 list pricing, and A...
SpaceXAI released Grok 4.5 on July 8, and for once the vendor hype and the third-party numbers point roughly the same direction. Musk called it "roughly comparable to Opus 4.7, but much faster." Priced at $2 per million input tokens and $6 per million output, that's over 60% b...
GPT-5.6 Luna went to $0.20 input / $1.20 output per million tokens on July 30. That's an 80% cut. Terra dropped 20%. Luna's input now undercuts Gemini 3.1 Flash-Lite ($0.25/$1.50) and sits at one-fifth of Claude Haiku 4.5's $1 input. Simon Willison covered the announcement and...
Moonshot exposes an Anthropic-compatible endpoint, so pointing Claude Code at K3 means setting the Anthropic base URL and supplying a Moonshot key. No new CLI, no config rewrite. Hosted at $3/$15 per Mtok, same tier as Claude Sonnet 4.6, and Artificial Analysis scores K3 at 57...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.