The Gemini API CLI (gemini-api) is an experimental preview (v0.4.x) Google released on 2026-10-19: a single Go binary that turns Gemini models, media generation, and managed agents into stateless shell commands. It is not a coding agent. Use it when your own code decides the next step, Antigravity CLI when you want an agent to plan, and the Gen AI SDK for application backends.
Cloudflare's own models come in two classes: the Clef family returns option probabilities (clef 27B, clef-flash 9B, clef-omni 30B-A3B), while Clairic returns typed values and intervals. All four bill input tokens only ($0.038–$0.24 per million) and charge nothing for output; the three Clef models ship Apache-2.0 weights, and Clairic is still a hosted preview. They beat Jev on decision benchmarks and trail it on reasoning-heavy ones.
Decision models replace an LLM's "generate some text and you write the parser" with "you list the options up front and the model returns a probability for each." After TypeSafe proposed System One on 2026-09-15, this category went from one vendor to more than a dozen: Jev 1.14, the OpenAI Decisions API, Perplexity Decider v1.2, Microsoft-Decision-1, Liquid d1-ultra, Base Labs mammoth-1, vLLM d3, StartLux, Fastino GLiDE, Celeris-1 Decision, Inception Mercury Decide, and Cloudflare Clef. The two leaderboards disagree almost completely — the community Jev Decision Index 0.4.1 crown goes to GLiDE V1.2 (71.5), while the independent Artificial Analysis Decision Index crown goes to d1-ultra (65.2) with Jev 1.14 at only 57.1. Decision models win on classification and tool selection, lose on knowledge and reasoning, and are broadly overconfident.
CVE-2021-28476 is a bug in Hyper-V's virtual network switch driver: a guest VM sends a crafted request, the host code follows an address inside it without checking, and the host crashes. Akamai claims two scenarios, a crash and code execution, but the public PoC only shows the crash; code execution was not publicly demonstrated. CVSS 9.9, patched May 2021. On the bounty table, what was publicly shown maps to the DoS cell (up to $15,000); the L1 guest-escape cell tops out at $250,000 and needs a working RCE exploit, but without an exploit that cell still has $200,000, $100,000 and $50,000 rows.
Apodex is a reasoning-first Mixture-of-Experts model family built by the Apodex team for complex long-horizon research, forecasting, and verifiable reasoning tasks. An agent runtime supplies file, data, code, and tool execution; the core API accepts text only.
Cohere's North series marks its first agentic coding model family. North Mini Code (30B total / 3B active sparse MoE) is purpose-built for code generation and automated development, with 256K context and free access via OpenRouter as `north-mini-code:free`.
OpenRouter lists long context, image input, and tools for Dots3-Note Preview; verified weight access and licensing remain necessary for self-hosting.
LFM2.5-2.6B is an on-device hybrid reasoning model; checkpoint and OpenRouter context limits differ, with license and data terms to check before deployment.
LongCat-Flash and LongCat-2.0 are Meituan open-weight MoE models. This guide explains dynamic compute, LSA, long context, tool use, and the difference between model specifications and hosted endpoints.
Nemotron covers text reasoning, multimodal understanding, and safety classification. Ultra, Super, Lightning, Nano Omni, and Content Safety differ in architecture, context, and licensing; model specifications and API limits must be checked separately.
pplx-embed is the embedding family Perplexity has been releasing since February 2026, in three lines: v1 single-vector (0.6B/4B, API at $0.004/$0.03 per 1M tokens), context embeddings (v1 plus a 9B v2 preview), and v2-late multi-vector (0.6B/9B, searches PDF page images directly). They solve different problems; v2 does not replace v1.
There is no 'never comment' school — only 'comment by exception' (Uncle Bob) versus 'comments are part of the design' (Ousterhout, antirez). In their 2024–2025 public debate, both agree on why-comments and against noise comments; the real fights are over interface comments for internal methods, long names as a substitute, and whether comments can be trusted. Controlled studies say quality decides: the same comments moved performance anywhere from -30% to +34% depending on the snippet.
Gemma is Google's open-weights family paired with the closed-source Gemini flagship; Gemma 4 (E2B / E4B / 26B-MoE / 31B) released in April 2026 switched its license from Google Gemma Terms of Use to Apache 2.0 — the single most important change in this article.
Laguna is Poolside's agentic coding model family: XS 2.1 packs 33B-A3B into a 36GB Mac, while S 2.1 brings 118B-A8B with 1M context to 70.2% on Terminal-Bench 2.1 and 40.4% on DeepSWE, both open under OpenMDW-1.1.
Ling separates execution, Ring handles deep reasoning, and Ming covers multimodal tasks; the family prioritizes intelligence per unit of compute, so selection starts with workload, deployment budget, and modality.
Muse Spark is Meta's closed-source agentic model line: version 1.3 combines a 1M-token context, multimodal inputs, and long-horizon tool loops. Standard pricing is $1.25/$4.25 per 1M input/output tokens, while Contributor drops to $0.10/$0.20 in exchange for training rights. It is not the next Llama; it is a separate product line built around models, APIs, and coding agents.
Nex-N2.5 is Nex AGI's open agentic model family: mini scores 82.9 on OSWorld-G at 35B-A3B, Pro tops Claude Opus 5 with 87.4 at 397B-A17B, and Max leads the whole official table on BrowseComp with 92.6 at 1.6T, all open under Apache-2.0.
GPUtw AI Skill is an official Agent Skill pack plus MCP server that lets Claude Code, Codex, Cursor, Copilot, and Gemini CLI manage GPUtw's Taiwan-based GPU cloud using natural language. 18 MCP tools cover catalog queries, instance lifecycle, status monitoring, and Vault operations. MIT licensed.
document_parse gains an effort parameter (fast/standard/deep) unifying anydoc WASM, OCR.space, and Docling VLM into a single dial; document_chunk's fieldAware mode extracts field names from table headers and metadata to solve RAG attribute conflation; document_smart_parse now returns a confidence score so agents decide whether to upgrade.
Groundlane v0.1.0 adds four quality mechanisms: web_extract selector healing (three-layer deterministic fallback, no LLM), corpus_retrieval_test (RAG recall verification inspired by RAGFlow), a document benchmark CLI (character-level F1), and a search benchmark framework (ground truth corpus with multi-provider comparison). Tool count goes from 54 to 55.