~/beer-and-code
▪ Clã Beer and Code the largest AI Engineering community in Brazil · live, every week join the Clã ›
~ / news $ ls -la

News

Releases, news, and trends from the world of software development.

77 news
13 #llm · #claude
Kimi K3.1 Pro vs Claude Opus 5.5: half the price, 0.7 points behind

Kimi K3.1 Pro lands 0.7 points behind Claude Opus 5.5 on Terminal-Bench 4.0 at half the price: $2/$10 vs $4/$20. We separate what is verified from what Moonshot self-reported, run the cost-per-task math (verbosity eats part of the savings), cover the quota burn and the weights promised for Nov 4, and show how to test both on your own workload.

26 Oct · 14 min ›
14 #claude · #anthropic
Anthropic S-1: up to 7% of revenue came from rule-breakers (and the $80B loss)

Anthropic's S-1 makes two confessions. The loud one is the $80.7B loss, almost all of it accounting. The quiet one: 5% to 7% of revenue came from customers who broke the usage policy, and Anthropic is going to cut them off. We read the filing separating what's in the S-1, what's press coverage and what's rumor, and what it means for your token price and your account.

26 Oct · 14 min ›
15 #embeddings · #pgvector
EmbeddingGemma 2: Google's open source embedding that searches text inside images

Google released EmbeddingGemma 2: an open source multimodal embedding, Apache 2.0, that puts images and text in the same vector space and runs on your phone. Here are the model card numbers, a screenshot indexer with pgvector, a comparison with Gemini Embedding 2, and where OCR is still mandatory.

19 Oct · 14 min ›
16 #agentes · #google
Harness2: Google open-sourced the agent that rewrites its own harness

Google open-sourced Harness2, a framework where the agent rewrites its own harness (prompt, tools, memory, skills, subagents) without touching the weights and without ground truth. What it edits, what the numbers actually prove, how to run it, and how to govern a harness that edits itself.

19 Oct · 13 min ›
17 #ia · #llm
Elorian Glimpse: the VLM that looks at the image again before it answers

Elorian Glimpse is a VLM that goes back to the image mid-reasoning: it crops, rewatches frames, and measures with code. Benchmarks, pricing ($0.60/M), and where to test it.

19 Oct · 14 min ›
18 #laravel · #google
Gemini 4 Argon vs Claude Opus 5.5: half the price until Nov 30, but is it worth switching?

Gemini 4 Argon beats Claude Opus 5.5 on 14 of 18 of Google's benchmarks and costs half the price. But it loses on Terminal-Bench, where coding agents live, and on December 1 the price evens out. The numbers side by side (Google's and independent ones), what an agentic session costs on all three with GPT-6.1 Sol as the yardstick, what breaks when you migrate, and a Laravel fallback switched by .env.

16 Oct · 12 min ›
19 #llm · #seguranca-ia
Anthropic accuses Moonshot of serving Claude instead of Kimi: what the report proves (and what it doesn't)

Anthropic accuses Moonshot of relaying Kimi customer requests to Claude Opus and showing the answer as if it came from Kimi: almost 300,000 requests in ten days, 5,380 fraudulent accounts and 23 million distillation exchanges. We separate what the 154-page report says, what it doesn't prove, what Moonshot has (not) answered and what changes for anyone running Kimi K3 in production.

06 Oct · 15 min ›
20 #claude · #anthropic
Claude Fable 5.5: What's Fact, What's Rumor, and Why the November IPO Entered the Conversation

Claude Fable 5.5 has not been announced: no model id, no model card, no pricing. A tracker, with dates and sources, of what's confirmed (Fable 5.1 is still the current model, Anthropic's IPO is targeted for mid-November) and what's just timeline chatter (hidden routing, viral demos, a URL returning 404). Plus what to do with your code in the meantime.

02 Oct · 10 min ›
21 #openai · #seguranca-ia
PewDiePie Banned by OpenAI Twice: What Distillation Is and What Ajax Has to Do With It

PewDiePie says OpenAI banned his account twice for distillation while he was training Ajax, a local Qwen 3.5 9B with refusals removed by Heretic. What model distillation is, how OpenAI detects it (the report on the Moonshot case came out two days earlier), what's inside Ajax, and where the line sits between legitimate synthetic data and a ban on your account.

02 Oct · 10 min ›
22 #openai · #alucinacao
GPT-6 Sol and Luna, Opus 5.5, and Gemini 4: What Actually Changes in This Wave of Models

GPT-6 Sol and Luna arrived at half the price of GPT-5.6, Claude Opus 5.5 delivers Fable 5.1 level at $4/$20 with four breaking changes in the API, and Google says Gemini 4 ships before the end of the year. What each one brings that's different, the catches (Luna regresses on agentic coding, Opus default effort dropped to medium), and which one to use now.

24 Sep · 13 min ›
23 #harness · #llm
Grok 4.7 Answers in 0.85s and Fable 5.1 in 298s: Where Each One Pays Off

Grok 4.7 landed on September 21, 2026 costing 5x less per token than Fable 5.1 and GPT-6 Astra, and it still loses to Astra on cost per completed task. Where this model actually pays off (spoiler: latency, not code), where it sits 22 points behind on Terminal-Bench, and the math that takes apart the list-price comparison.

22 Sep · 14 min ›
24 #modelos-de-ia · #custos
Laya vs Jev: the US$40 million decision AI against the free one that runs on your laptop

Jev costs US$0.042 per million tokens and is a closed API. Laya is Apache 2.0, runs offline on a 2018 GPU and measures 7.8x faster at P50. A comparison with the numbers each side published, the prior-art fight that broke out on Hacker News, and the October update: Cloudflare's open-weight Clef and OpenAI's Decisions API on GPT-6 Luna.

21 Sep · 21 min ›
Meet the Clã Beer and Code
playing