Release watch
New AI model releases this week and the last 30 days
What arrived, what got measured, and who is due. Rebuilt once a day from the day’s data.
In the week to 6 Nov 2026, 3 AI models were released and 1 already has independent results: Inkling 2 Small (Thinking Machines, 5 Nov) is the highest-ranked new model at 57.4. 7 further models are in provider catalogues without independent scores yet. BenchLeader tracks 24 AI models released in October 2026; the highest-ranked is Gemini 4 Carbon (69.2).
Key facts (as of 6 Nov 2026)
- Thinking Machines released Inkling 2 Small on 5 Nov 2026 (first seen by BenchLeader 6 Nov 2026).
- Inkling 2 Small scores 57.4 on the BenchLeader Index, from independent results only (BenchLeader, 6 Nov 2026).
- Inkling 2 Small costs $0.475 per 1M tokens blended (list price, 6 Nov 2026).
- BenchLeader lists a new model the day a provider's API or catalogue shows it, and ranks it once it has 5 independent results across 2 categories (BenchLeader methodology).
Data as of 6 Nov 2026
New this week
| Model | Released | Index | Rank | Price $/M | Status |
|---|---|---|---|---|---|
| Inkling Small 2Unknown | seen 6 Nov 2026 | – | – | – | First results arriving |
| Inkling 2 SmallThinking Machines | 5 Nov 2026 | 57.4 | #146 | $0.475 | Ranked |
| Crouton AlphaUnknown | seen 5 Nov 2026 | – | – | – | Listed; awaiting independent results |
| seen 5 Nov 2026 | – | – | $3.38 | Listed; awaiting independent results | |
| seen 5 Nov 2026 | – | – | $3.38 | Listed; awaiting independent results | |
| Ripple-65B PreviewRicursive | 4 Nov 2026 | – | – | $0.300 | Released; awaiting independent results |
| 2 Nov 2026 | – | – | – | First results arriving | |
| seen 2 Nov 2026 | – | – | $0.590 | Listed; awaiting independent results |
Released in the last 30 days
Every model released, or first listed with no release date, in the 30 days to 6 Nov 2026, newest first.
Who is due
Expected windows from each provider’s release cadence over the last two years. An estimate from dates only, not from announcements.
| Provider | Last release | Cadence | Expected window |
|---|---|---|---|
| 30 Apr 2025 | every 62 days | 6 Jun 2025 to 25 Jul 2025overdue | |
| 12 Dec 2025 | every 112 days | 17 Feb 2026 to 17 May 2026overdue | |
| 2 Dec 2025 | every 148 days | 28 Feb 2026 to 27 Jun 2026overdue | |
| 8 Jan 2026 | every 93 days | 4 Mar 2026 to 18 May 2026overdue | |
| 21 Jul 2026 | every 65 days | 29 Aug 2026 to 20 Oct 2026overdue | |
| 17 Sep 2026 | every 7 days | 21 Sep 2026 to 26 Sep 2026overdue | |
| 29 Jun 2026 | every 152 days | 28 Sep 2026 to 27 Jan 2027 | |
| 28 Aug 2026 | every 53 days | 28 Sep 2026 to 10 Nov 2026 | |
| 18 Sep 2026 | every 28 days | 4 Oct 2026 to 27 Oct 2026overdue | |
| 21 Sep 2026 | every 26 days | 6 Oct 2026 to 27 Oct 2026overdue | |
| 1 Oct 2026 | every 19 days | 12 Oct 2026 to 27 Oct 2026overdue | |
| 21 Sep 2026 | every 41 days | 15 Oct 2026 to 17 Nov 2026 | |
| 22 Sep 2026 | every 41 days | 16 Oct 2026 to 18 Nov 2026 | |
| 7 Oct 2026 | every 24 days | 21 Oct 2026 to 9 Nov 2026 | |
| 18 Sep 2026 | every 59 days | 23 Oct 2026 to 9 Dec 2026 | |
| 9 Sep 2026 | every 76 days | 24 Oct 2026 to 24 Dec 2026 | |
| 25 Aug 2026 | every 119 days | 4 Nov 2026 to 7 Feb 2027 | |
| 28 Oct 2026 | every 15 days | 6 Nov 2026 to 18 Nov 2026 | |
| 15 Oct 2026 | every 40 days | 8 Nov 2026 to 10 Dec 2026 | |
| 17 Oct 2026 | every 37 days | 8 Nov 2026 to 7 Dec 2026 | |
| 26 Oct 2026 | every 26 days | 10 Nov 2026 to 1 Dec 2026 | |
| 19 Oct 2026 | every 38 days | 10 Nov 2026 to 11 Dec 2026 | |
| 23 Oct 2026 | every 36 days | 13 Nov 2026 to 12 Dec 2026 | |
| 29 Oct 2026 | every 28 days | 14 Nov 2026 to 7 Dec 2026 | |
| 19 Oct 2026 | every 57 days | 22 Nov 2026 to 6 Jan 2027 | |
| 23 Oct 2026 | every 56 days | 25 Nov 2026 to 9 Jan 2027 | |
| 2 Nov 2026 | every 73 days | 15 Dec 2026 to 12 Feb 2027 |
First results this week
Benchmarks that published a first score for a model in the last seven days.
- DeepSeek V4.1 Flash 1017
- ARC-AGI-3: 5.8%3 Nov 2026
- AA Intelligence Index v5.0: 28.52 Nov 2026
- AA Progress Index v1: 319.32 Nov 2026
- HLE-Diamond (AA): 20.8%2 Nov 2026
- MMMU-Pro: 79.7%2 Nov 2026
- AA-Omniscience v1.1: 1.52 Nov 2026
- and 28 more
- Muse Wonder
- AA Intelligence Index v5.0: 40.55 Nov 2026
- HLE-Diamond (AA): 34.5%5 Nov 2026
- MMMU-Pro: 83.8%5 Nov 2026
- AA-Omniscience v1.1: 24.55 Nov 2026
- CritPt v2 hard subset (AA): 51%5 Nov 2026
- Terminal-Bench 4.0 (AA): 38.4%5 Nov 2026
- and 8 more
- Hy4
- AA Intelligence Index v5.0: 34.65 Nov 2026
- HLE-Diamond (AA): 32.7%5 Nov 2026
- AA-Omniscience v1.1: 5.45 Nov 2026
- CritPt v2 hard subset (AA): 39%5 Nov 2026
- Terminal-Bench 4.0 (AA): 37.4%5 Nov 2026
- AA-Omniscience v1.1: accuracy: 50%5 Nov 2026
- and 6 more
- Claude Opus 5
- AA Intelligence Index v5.0: 44.25 Nov 2026
- HLE-Diamond (AA): 44.8%5 Nov 2026
- AA-Omniscience v1.1: 37.65 Nov 2026
- CritPt v2 hard subset (AA): 50%5 Nov 2026
- AA-Omniscience v1.1: accuracy: 60.9%5 Nov 2026
- AA-Omniscience v1.1: non-hallucination: 40.6%5 Nov 2026
- and 2 more
- Grok 4.6
- AA Intelligence Index v5.0: 315 Nov 2026
- HLE-Diamond (AA): 21.5%5 Nov 2026
- Terminal-Bench-Science 0.1 (AA): 2.9%5 Nov 2026
- AA-Omniscience v1.1: 24.94 Nov 2026
- CritPt v2 hard subset (AA): 40%4 Nov 2026
- AA-Omniscience v1.1: accuracy: 46.2%4 Nov 2026
- and 2 more
- Gemini 3.1 Flash Lite
- AA Intelligence Index v5.0: 6.25 Nov 2026
- CritPt v2 hard subset (AA): 3%5 Nov 2026
- FrontierCode 2.0 (AA): 0.6%5 Nov 2026
- HLE-Diamond (AA): 9%4 Nov 2026
- AA-Omniscience v1.1: -24.14 Nov 2026
- AA-Omniscience v1.1: accuracy: 31.7%4 Nov 2026
- and 2 more
- North Mini Code
- AA Intelligence Index v5.0: 3.46 Nov 2026
- FrontierCode 2.0 (AA): 0.2%6 Nov 2026
- Terminal-Bench-Science 0.1 (AA): 0%6 Nov 2026
- AA-Omniscience v1.1: -49.85 Nov 2026
- CritPt v2 hard subset (AA): 1%5 Nov 2026
- AA-Omniscience v1.1: accuracy: 18.4%5 Nov 2026
- and 1 more
- GPT-5.6 Luna
- AA Intelligence Index v5.0: 23.85 Nov 2026
- HLE-Diamond (AA): 20.9%5 Nov 2026
- CritPt v2 hard subset (AA): 30.5%5 Nov 2026
- AA-Omniscience v1.1: -17.14 Nov 2026
- AA-Omniscience v1.1: accuracy: 39.1%4 Nov 2026
- AA-Omniscience v1.1: non-hallucination: 7.7%4 Nov 2026
- and 1 more
- Gemma 4 31B
- AA Intelligence Index v5.0: 4.45 Nov 2026
- HLE-Diamond (AA): 7.2%5 Nov 2026
- FrontierCode 2.0 (AA): 0%5 Nov 2026
- Terminal-Bench-Science 0.1 (AA): 0%5 Nov 2026
- AA-Omniscience v1.1: -48.530 Oct 2026
- AA-Omniscience v1.1: accuracy: 19.9%30 Oct 2026
- and 1 more
- Claude Haiku 4.5
- AA Intelligence Index v5.0: 7.45 Nov 2026
- CritPt v2 hard subset (AA): 1.5%5 Nov 2026
- FrontierCode 2.0 (AA): 0.4%5 Nov 2026
- AA-Omniscience v1.1: -1.84 Nov 2026
- AA-Omniscience v1.1: accuracy: 17.5%4 Nov 2026
- AA-Omniscience v1.1: non-hallucination: 76.6%4 Nov 2026
- and 1 more
- Granite 4.2 3B
- AA Intelligence Index v5.0: 4.731 Oct 2026
- FrontierCode 2.0 (AA): 0%31 Oct 2026
- Terminal-Bench-Science 0.1 (AA): 0%31 Oct 2026
- AA-Omniscience v1.1: -18.330 Oct 2026
- CritPt v2 hard subset (AA): 0%30 Oct 2026
- AA-Omniscience v1.1: accuracy: 9.2%30 Oct 2026
- and 1 more
- Solar Pro 4
- HLE-Diamond (AA): 4.7%6 Nov 2026
- AA-Omniscience v1.1: 1.46 Nov 2026
- CritPt v2 hard subset (AA): 18%6 Nov 2026
- AA-Omniscience v1.1: accuracy: 18.4%6 Nov 2026
- AA-Omniscience v1.1: non-hallucination: 79.1%6 Nov 2026
- Terminal-Bench-Science 0.1 (AA): 0%5 Nov 2026
- GPT-6 Sol
- AA Intelligence Index v5.0: 44.14 Nov 2026
- AA-Omniscience v1.1: 21.44 Nov 2026
- CritPt v2 hard subset (AA): 66%4 Nov 2026
- AA-Omniscience v1.1: accuracy: 51.1%4 Nov 2026
- AA-Omniscience v1.1: non-hallucination: 39.2%4 Nov 2026
- FrontierCode 2.0 (AA): 35.6%4 Nov 2026
- Granite 4.2 30B
- Terminal-Bench-Science 0.1 (AA): 0%2 Nov 2026
- CritPt v2 hard subset (AA): 1.5%31 Oct 2026
- HLE-Diamond (AA): 4.3%30 Oct 2026
- AA-Omniscience v1.1: -12.430 Oct 2026
- AA-Omniscience v1.1: accuracy: 11.2%30 Oct 2026
- AA-Omniscience v1.1: non-hallucination: 73.3%30 Oct 2026
- Granite 4.2 8B
- AA Intelligence Index v5.0: 5.12 Nov 2026
- FrontierCode 2.0 (AA): 0%2 Nov 2026
- AA-Omniscience v1.1: -16.731 Oct 2026
- AA-Omniscience v1.1: accuracy: 11.8%31 Oct 2026
- AA-Omniscience v1.1: non-hallucination: 67.8%31 Oct 2026
- Terminal-Bench-Science 0.1 (AA): 0%31 Oct 2026
- Qwen3.8 27B
- AA Intelligence Index v5.0: 1031 Oct 2026
- FrontierCode 2.0 (AA): 3.7%31 Oct 2026
- AA Intelligence Index v5.0: 17.530 Oct 2026
- FrontierCode 2.0 (AA): 11.5%30 Oct 2026
- AA Intelligence Index v5.0: 15.730 Oct 2026
- FrontierCode 2.0 (AA): 12.4%30 Oct 2026
- DeepSeek V4 Pro
- AA Intelligence Index v5.0: 12.65 Nov 2026
- CritPt v2 hard subset (AA): 8.5%5 Nov 2026
- Terminal-Bench-Science 0.1 (AA): 0%5 Nov 2026
- FrontierCode 2.0 (AA): 10.3%4 Nov 2026
- MirrorCode: 33.3%31 Oct 2026
- Nemotron 3.5 Lightning
- AA Intelligence Index v5.0: 5.530 Oct 2026
- AA-Omniscience v1.1: -15.730 Oct 2026
- AA-Omniscience v1.1: accuracy: 14.7%30 Oct 2026
- AA-Omniscience v1.1: non-hallucination: 64.3%30 Oct 2026
- FrontierCode 2.0 (AA): 0%30 Oct 2026
- Kimi K3
- FACTS Search: 0.9%5 Nov 2026
- MirrorCode: 54.4%31 Oct 2026
- AA Intelligence Index v5.0: 17.430 Oct 2026
- FrontierCode 2.0 (AA): 10.9%30 Oct 2026
- GLM 5.2
- FORTRESS: 48.7%5 Nov 2026
- GameBench 2 by reasoning setting: 32.94 Nov 2026
- GameBench 2 by reasoning setting: 51.62 Nov 2026
- GameBench 2 by reasoning setting: 442 Nov 2026
- Mistral Medium 3.5
- AA Intelligence Index v5.0: 4.75 Nov 2026
- CritPt v2 hard subset (AA): 0.5%5 Nov 2026
- FrontierCode 2.0 (AA): 1.8%5 Nov 2026
- Terminal-Bench-Science 0.1 (AA): 0%30 Oct 2026
- Seed 1.6
- Chess Puzzles: 5%5 Nov 2026
- GPQA Diamond: 77.6%4 Nov 2026
- OTIS Mock AIME: 69.4%4 Nov 2026
- SimpleQA Verified: 17.4%4 Nov 2026
- Claude Fable 5.1
- GameBench 2 by reasoning setting: 94.14 Nov 2026
- GameBench 2 by reasoning setting: 82.94 Nov 2026
- GameBench 2 by reasoning setting: 84.84 Nov 2026
- MirrorCode: 93.3%31 Oct 2026
- GPT-5.6 Terra
- AA Intelligence Index v5.0: 32.34 Nov 2026
- CritPt v2 hard subset (AA): 50.5%4 Nov 2026
- FrontierCode 2.0 (AA): 23.9%4 Nov 2026
- MirrorCode: 41.1%31 Oct 2026
Most-searched head-to-heads
The model comparisons people look up most, each with a verdict, category wins, price and speed.
Questions people ask
- What AI models were released this week?
- In the seven days to 6 Nov 2026: Inkling Small 2 (Unknown, 6 Nov 2026), Inkling 2 Small (Thinking Machines, 5 Nov 2026), Crouton Alpha (Unknown, 5 Nov 2026), Gemini Omni 1.1 Flash (Google, 5 Nov 2026), Gemini Omni 1.2 Flash (Google, 5 Nov 2026), Ripple-65B Preview (Ricursive, 4 Nov 2026), Doubao Seed Evolving (ByteDance, 2 Nov 2026) and GLM 5.3 FlashX Legal (Zhipu AI, 2 Nov 2026).
- What new AI models came out today?
- Today's data run (6 Nov 2026) added Inkling Small 2.
- What is the best new AI model this month?
- Of the models released in the 30 days to 6 Nov 2026, Gemini 4 Carbon (high effort) by Google ranks highest, at 69.2 on the BenchLeader Index (top 10 overall).
- Why does a new model have no score yet?
- The BenchLeader Index needs at least five independent results across at least two categories, and scores from a model's own launch post are never used. A new model is listed straight away and ranked once independent publishers have measured it, usually within days to weeks.
- Which AI labs are due to release next?
- Going by release cadence alone, not announcements: Meituan (between 28 Sep 2026 and 27 Jan 2027), Tencent (between 28 Sep 2026 and 10 Nov 2026) and SpaceXAI (between 15 Oct 2026 and 17 Nov 2026). The table below has every lab.