AI releases, day by day

Every model and product release the daily digest has listed under “Notable AI releases”, newest first. The score and price on a line are as they stood on launch day, and are blank where none was available that day. Each date links to that day’s edition.

November 2026

12 Nov · Gemini Robotics 2.5 · thread · agent — DeepMind's top robot-control model; 30–100% higher task success than earlier baselines across 200 tasks (self-reported); rolling out to robotics partners only.
12 Nov · Gemini Decision 1 · thread · small/cheap · $0.10 in per MTok — New kind of Gemini model for closed-set decisions: scores up to 1,000 yes/no, choice or rubric questions in one pass over text, images, audio, video and PDFs.
12 Nov · LongCat-2.5 · open weights — Meituan's 1.6-trillion-parameter multimodal agent model (about 48B active per token, 1M-token context) under an MIT licence; benchmark claims are self-reported.
12 Nov · AlphaGenome 2 · thread · product — Domain-specific DNA model predicting how variants act in specific cell types; API open for non-commercial research, commercial use in private preview.
11 Nov · Gemini 4 Carbon · thread · frontier · $2 in / $10 out per MTok — General availability after a cyber-defender-only preview; second on Artificial Analysis's index at a quarter of Fable 5.2's per-token price
10 Nov · Claude Fable 5.2 · thread · frontier · $8 in / $40 out per MTok — New leader on Artificial Analysis's Intelligence Index (65, vs Gemini 4 Carbon's 60), and 20% cheaper per token than Fable 5.1
10 Nov · Claude Mythos 5.2 · thread · frontier · $8 in / $40 out per MTok — Same model as Fable 5.2 with reduced cyber and biology safeguards; vetted access only, via Anthropic's Cyber and Life Sciences Verification Programs
10 Nov · Muse Ember · small/cheap · $0.10 in / $0.50 out per MTok — Meta's smallest model; Meta says it matches Claude Haiku 5.5's best Artificial Analysis score at about a third of the cost per task, though it trails on coding
9 Nov · Cola2 · open weights — Experimental 9B model from ByteDance Seed that generates text through continuous latents and a diffusion transformer rather than token by token; benchmark results preliminary
7 Nov · Inkling 2 Small · open weights — Highest-scoring US open-weight model on the Vals Index and #7 among open-weight models globally, per Vals; uses about 5x the tokens of its predecessor
7 Nov · ERNIE 5.5 Flash Base · open weights — incremental; a base (not instruction-tuned) model from Baidu
6 Nov · GPT-6 Cyber · thread · frontier — Cyber-tuned GPT-6 Astra rated Critical for cybersecurity, gated to vetted Daybreak Red customers; ExploitGym 48.5% vs 32.1% for GPT-5.6 Cyber
6 Nov · GPT-Image-2.6 Sunburst · thread · image — OpenAI's higher-quality new image model, with better consistency across edits; incremental
6 Nov · GPT-Image-2.6 Flare · thread · image — Faster sibling, generating images in as little as six seconds; incremental
6 Nov · Gemini Omni 1.2 Flash · video — Google video model point update: 15-second clips, multi-turn editing, 30% cheaper than Omni Flash 1.1; incremental
5 Nov · Hy4 · mid-tier — Full release after September's open-weights preview, adding vision and long-horizon agent features; Tencent says weights and system cards follow in days.
5 Nov · Kimi K3.1 Pro (open weights) · thread · open weights — Weights for Moonshot's 2.8T-parameter MoE (104B active), 4-bit, custom licence; API launch was 24 Oct. Moonshot says it removed cyber training data for the open release.
1 Nov · GPT-Rosalind-6 · frontier — GPT-6 Astra weights with fewer refusals on dual-use biology, available only to vetted researchers; OpenAI rates it High in bio/chem, inherits Astra's Critical cyber rating

October 2026

30 Oct · Mistral Large 4 (Le Chonk) · open weights — 1T-parameter multimodal model now open; per Artificial Analysis the top open-weight model from outside China, though below the leading Chinese open models
29 Oct · Gemini 4 Carbon (preview) · thread · frontier · AA Intelligence Index 59 · $0.10 cached / $2 in / $10 out per MTok — New #1 on AA Intelligence Index v5; limited at first to selected cyber defenders; introductory price is half the $4/$20 standard rate
28 Oct · Genie 4 · thread · video — DeepMind's most realistic real-time interactive world model; adds native audio and keeps scenes consistent for 2 minutes. Trusted testers only for now, with Ultra and API access to follow
28 Oct · MAI Image 2.7 (preview) · image — Near the top for image editing: Microsoft puts it No. 2 on Artificial Analysis, 6 Elo behind GPT Image 2.5 Sunburst. Playground and Arena only; no API or price yet
27 Oct · GPT-6.1 Terra · thread · mid-tier · $0.05 cached / $1 in / $5 out per MTok — New cost-efficiency leader per Artificial Analysis: 6 index points above GPT-5.6 Terra at $0.40 vs $1.40 per task, 4 points below GPT-6.1 Sol; hallucination rate 88%→38%
25 Oct · Dessa · speech — Block's full-duplex speech-to-speech model, trained from scratch only on human conversation audio with no text; runs locally on a high-end MacBook. Quality claims are Block's own.
24 Oct · Kimi K3.1 Pro · frontier · AA Intelligence Index 55 · $0.20 cached / $2 in / $10 out per MTok — Fourth on the AA index. It becomes the top open-weights model once weights ship by 4 Nov, cutting the open–closed gap to 3 points.
22 Oct · RFD4-Proteina · thread · open weights — NVIDIA and UW's Institute for Protein Design follow up RFdiffusion3, claiming 3.8x faster generation. The code is on GitHub; the weights sit in a gated Hugging Face repo.
20 Oct · MiniMax M3.1 Flash · small/cheap · AA Intelligence Index 42 · $0.015 cached / $0.15 in / $0.60 out per MTok — Strong cheap agentic model: 39.3% on PostTrainBench against Opus 5's 40.4% (per Maksym Andriushchenko); MiniMax says weights will follow later
20 Oct · Mercury 3 · small/cheap · $0.02 cached / $0.20 in / $0.70 out per MTok — Inception claims the most capable diffusion LLM yet, matching GPT-6 Luna and Claude Haiku 5.5 at 1,000+ tokens/s; benchmarks self-reported
18 Oct · DeepSeek-V4.1-Flash-1017 · open weights — Update checkpoint of DeepSeek's cheap Flash tier, with self-reported coding and agent gains and shorter runs; now on the API, web and app.
18 Oct · FLUX 3 Image Klein · image · open weights — 5.6B open-weights sibling of FLUX 3 Image: text-to-image, editing and multi-reference generation with up to 10 images, at native 4K.
17 Oct · Beam · open weights · AA Intelligence Index 30 — Reflection AI's 501B open model ranks second among US open-weight models on Artificial Analysis, which also reports 34% on its cyber index
16 Oct · Muse Spark Open · open weights — Strongest US open-weight model: 816B MoE (42B active), 1M context, runs on two Mac Studios. Meta's own table gives it AA 39, behind GLM 5.3 (45) and Kimi K3 (44).
16 Oct · Nemotron 3.5 Super VL · thread · open weights · AA Intelligence Index 26.5 — 120B vision-language model. Doubles its predecessor's AA score while using about 60% fewer output tokens per task.
16 Oct · Nemotron 3.5 Bolt 74B-A9B · thread · small/cheap · AA Intelligence Index 21 · open weights — Fits in 24GB for local agents. Scores within 2 points of the 550B Nemotron 3 Ultra.
16 Oct · GPT-Live-1.1 · thread · voice — Update to OpenAI's full-duplex voice model in ChatGPT, Codex and dots: 13 new voices, better non-English accents. No benchmarks yet; GPT-Live-1 trails Grok Voice 3.0 on Artificial Analysis's speech index.
15 Oct · Grok Voice 3.0 · speech · $0.08 per minute — xAI says it ranks first on Artificial Analysis's Speech-to-Speech Index (86.9); same price as Grok Voice 2.0
14 Oct · Gemini 4 Argon (general availability) · thread · frontier — Moves from a cyber-defender-only preview to the Gemini API and app. Google cannot rule out that it has reached its cyber and bio critical capability levels.
13 Oct · Muse Wonder Full · frontier — Meta's first top-tier model, second only to Claude Opus 5.5 on Artificial Analysis's index; Muse Code subscribers only, no API yet
13 Oct · Muse Wonder · frontier — API version just below the top tier; big agentic-coding gains over Muse Spark 1.3
13 Oct · MiMo-V2.6 Pro · open weights — released with an RL training framework and 7,000+ task environments, including cyber and software-engineering tasks
13 Oct · MiMo-V2.6 Flash · open weights — incremental; cheaper sibling of V2.6 Pro, alongside a distilled 9B model
13 Oct · GLM-Legal · open weights — domain-specific; Vals AI ranks it the top open-weight model on Harvey's Legal Agent Benchmark and LegalBench
11 Oct · Step 5 Preview · mid-tier — StepFun's 1M-token-context mixture-of-experts model, available as a preview on OpenRouter; tier, pricing and weights status not yet published
10 Oct · GPT-6 Sol and GPT-6 Luna (October update) · frontier — Update, not a new model: October versions are now the default for all ChatGPT users, free tier included; OpenAI rates both High in cyber and bio, with regressions on some content-safety evals.
9 Oct · EmbeddingGemma 2 · open weights — Google DeepMind's open, lightweight embedding model, now multimodal; incremental
8 Oct · Claude Haiku 5.5 · thread · small model · AA Intelligence Index 43 (max effort) · $0.01 / $0.10 / $0.50 per MTok (prompts up to 100k tokens; $0.05 / $0.50 / $2.50 — Anthropic's new small model, about 75% cheaper to run than Haiku 4.5. It scores 72.4% on OSWorld 2.1 (offline subset), against 15.7% for Haiku 4.5, and is the first Haiku with an adjustable effort setting.
7 Oct · Mistral Large 4 · mid-tier · $1.36 in / $4.18 out per MTok — Mistral's largest model (1T params, 49B active), in API preview. Weights are promised by end of October after cyber red-teaming with partners and state authorities.
7 Oct · Nano Banana 2.1 · thread · image model — Google's updated Flash-tier image generation and editing model, built on Gemini 3.6 Flash. It improves visual design, mask-based editing and subject consistency, and is live in the Gemini app, Search AI Mode and the Gemini API.
1 Oct · Gemini 4 Argon · thread · frontier · $2 in / $10 out per MTok — New #1 on the Vals Index; restricted preview for vetted cyber defenders, who get it without cyber guardrails; the price is introductory
1 Oct · dots · thread · agent — Always-on GPT-6 Astra agents, each with its own cloud computer and access to 4,000+ connected apps, for Pro, Business Premium and Enterprise users

September 2026

30 Sep · GPT-6.1 Sol · thread · frontier · AA Intelligence Index one point below GPT-6 Astra · $0.10 cached / $2 in / $10 out per MTok — Near-Astra intelligence at ~a quarter of the cost per task ($0.72 vs $3.26); pushes out the cost-efficiency frontier. System card rates it critical in cyber and high in bio/chem, with Astra-level safeguards.
30 Sep · Quine · product — Early-stage Microsoft Research system building a multimodal 'world model' of biology to prioritise hypotheses before the lab; research preview, not a general model.
29 Sep · Claude Sonnet 5.5 · thread · mid-tier · $0.20 cached / $2 in / $10 out per MTok — Scores 70.6% on Terminal-Bench 4.0, up from Sonnet 5's 10.3%. First Sonnet launched with Opus-class cyber safeguards. Vals ranks it #2, just behind Opus 5.5.
25 Sep · GPT-6 Luna · small/cheap — OpenAI's cheap tier: Vals lists it at $0.10/$0.50 per MTok, about 100x below GPT-6 Astra, and within 8 points of Astra on the Vals Index (Released on 23 September 2026)
24 Sep · Gemini 3.8 Flash TTS · thread · speech — Google's most expressive TTS yet, with voice design and voice replication in 100 languages; Google claims the top spot on Hume AI's voice benchmarks
24 Sep · Gemini 3.8 Flash-Lite TTS · thread · speech — Lighter sibling of Flash TTS in the same launch
24 Sep · Qwen-Image-2.1 · image · open weights — 7B open-weight image model that reportedly tops open-model arena categories, though early head-to-heads put it behind Google's Nano Banana 2
23 Sep · Claude Opus 5.5 · thread · frontier · AA Intelligence Index 58 · $0.20 cached / $4 in / $20 out per MTok — New top score on the Artificial Analysis Intelligence Index (58 vs 53 for GPT-6 Astra and Claude Fable 5.1) at 20% below Opus 5 pricing; first release since Anthropic's pacing call.
23 Sep · GPT-6 Sol · thread · mid-tier · AA Intelligence Index 48 · $2 in / $10 out per MTok — Level with GPT-5.6 Sol on the Intelligence Index (48 vs 47) at half the price; publishes alignment evals showing lower coding-deception rates than its predecessor.

← All past issues