Release watch
What arrived, what got measured, and who is due. Rebuilt every morning from the day’s data.
Data as of 5 Oct 2026
New this week
| Model | Released | Index | Rank | Price $/M | Status |
|---|---|---|---|---|---|
 Ling 3.1 FlashAnt Group | seen 3 Oct 2026 | – | – | – | Listed; awaiting independent results |
 Mercury VoiceInception | 1 Oct 2026 | – | – | $0.675 | Released; awaiting independent results |
 Gemini 4 ArgonhighGoogle | 30 Sept 2026 | 69.7 | #7 | $4.00 | Ranked |
 GPT-6.1 SolmaxOpenAI | 29 Sept 2026 | 69.5 | #11 | $4.00 | Ranked |
 Claude Opus 5.5 RangeAnthropic | seen 29 Sept 2026 | – | – | – | Listed; awaiting independent results |
| Claude Sonnet 5 RangeAnthropic | seen 29 Sept 2026 | – | – | – | Listed; awaiting independent results |
| Resonant 1Unknown | seen 29 Sept 2026 | – | – | – | Listed; awaiting independent results |
| Claude Sonnet 5.5xhighAnthropic | 28 Sept 2026 | 67.0 | #27 | $4.00 | Ranked |
| Daybreak BluemaxOpenAI | 7 Aug 2026 | – | – | $8.00 | Listed; awaiting independent results |
| Daybreak RedOpenAI | 7 Aug 2026 | – | – | $28.13 | Listed; awaiting independent results |
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 |
|---|---|---|---|
 Amazon | 2 Dec 2025 | every 148 days | 28 Feb 2026 to 27 Jun 2026overdue |
 Mistral AI | 29 Apr 2026 | every 44 days | 25 May 2026 to 29 Jun 2026overdue |
 ByteDance | 14 Feb 2026 | every 178 days | 31 May 2026 to 21 Oct 2026 |
 MiniMax | 1 Jun 2026 | every 108 days | 4 Aug 2026 to 30 Oct 2026 |
 Ai2 | 13 Jan 2026 | every 364 days | 19 Aug 2026 to 6 Jun 2027 |
 Moonshot AI | 16 Jul 2026 | every 62 days | 22 Aug 2026 to 10 Oct 2026 |
 Poolside | 21 Jul 2026 | every 65 days | 29 Aug 2026 to 20 Oct 2026 |
 NVIDIA | 11 Aug 2026 | every 34 days | 31 Aug 2026 to 27 Sept 2026overdue |
 Meta | 2 Sept 2026 | every 27 days | 18 Sept 2026 to 9 Oct 2026 |
 Alibaba | 17 Sept 2026 | every 7 days | 21 Sept 2026 to 26 Sept 2026overdue |
 DeepSeek | 10 Sept 2026 | every 20 days | 22 Sept 2026 to 8 Oct 2026 |
| Ant Group | 6 Aug 2026 | every 82 days | 24 Sept 2026 to 28 Nov 2026 |
 Zhipu AI | 18 Sept 2026 | every 23 days | 1 Oct 2026 to 20 Oct 2026 |
| Google | 30 Sept 2026 | every 15 days | 9 Oct 2026 to 21 Oct 2026 |
| OpenAI | 29 Sept 2026 | every 20 days | 11 Oct 2026 to 27 Oct 2026 |
 SpaceXAI | 21 Sept 2026 | every 42 days | 16 Oct 2026 to 18 Nov 2026 |
| Anthropic | 28 Sept 2026 | every 39 days | 21 Oct 2026 to 21 Nov 2026 |
| Inception | 1 Oct 2026 | every 34 days | 21 Oct 2026 to 17 Nov 2026 |
 Cohere | 9 Sept 2026 | every 87 days | 31 Oct 2026 to 8 Jan 2027 |
 IBM | 25 Aug 2026 | every 118 days | 3 Nov 2026 to 6 Feb 2027 |
 Xiaomi | 21 Sept 2026 | every 92 days | 15 Nov 2026 to 27 Jan 2027 |
 Upstage | 6 Aug 2026 | every 191 days | 28 Nov 2026 to 30 Apr 2027 |
 StepFun | 18 Sept 2026 | every 120 days | 29 Nov 2026 to 5 Mar 2027 |
 OpenBMB | 25 May 2026 | every 491 days | 15 Mar 2027 to 11 Apr 2028 |
First results this week
Benchmarks that published a first score for a model in the last seven days.
- GPT-6 Sol
- GPQA Diamond: 94.3%2 Oct 2026
- FrontierMath Tiers 1–3: 89.8%2 Oct 2026
- FrontierMath Tier 4: 90%2 Oct 2026
- OTIS Mock AIME: 100%2 Oct 2026
- SimpleQA Verified: 60.7%2 Oct 2026
- Mystery Game Puzzles: 56%2 Oct 2026
- and 38 more
- Claude Opus 5.5
- SciCode: 60.4%5 Oct 2026
- CursorBench: 56%5 Oct 2026
- ALE-Bench: 2147.45 Oct 2026
- SciCode: 59.3%5 Oct 2026
- APEX-Agents: 52.5%5 Oct 2026
- CursorBench: 52.5%5 Oct 2026
- and 28 more
- Grok 4.7
- GPT-6 Luna
- GPQA Diamond: 90.5%2 Oct 2026
- FrontierMath Tiers 1–3: 79%2 Oct 2026
- FrontierMath Tier 4: 56.1%2 Oct 2026
- OTIS Mock AIME: 98.9%2 Oct 2026
- SimpleQA Verified: 41.4%2 Oct 2026
- Chess Puzzles: 31%2 Oct 2026
- and 20 more
- Claude Sonnet 5.5
- LMArena Text: 14713 Oct 2026
- LMArena Hard Prompts: 15043 Oct 2026
- LMArena Coding: 15363 Oct 2026
- LMArena Vision: 12903 Oct 2026
- LMArena Creative Writing: 14473 Oct 2026
- LMArena Instruction Following: 14843 Oct 2026
- and 15 more
- Kimi K2.6
- AA Intelligence Index: 273 Oct 2026
- GPQA Diamond (AA): 91.1%3 Oct 2026
- Humanity's Last Exam (AA): 37.5%3 Oct 2026
- IFBench: 76%3 Oct 2026
- AA-LCR: 81%3 Oct 2026
- MMMU-Pro: 79.4%3 Oct 2026
- and 14 more
- Step 5 Preview
- ProofBench: 42%5 Oct 2026
- LMArena Text: 14543 Oct 2026
- LMArena Hard Prompts: 14823 Oct 2026
- LMArena Coding: 15113 Oct 2026
- LMArena Vision: 12783 Oct 2026
- LMArena Agent: 0.53 Oct 2026
- and 9 more
- Gemini 3.1 Pro
- PRBench Finance: 41.9%4 Oct 2026
- MultiChallenge: 71.4%3 Oct 2026
- PRBench Finance: 41.9%2 Oct 2026
- MultiNRC: 64.7%2 Oct 2026
- SWE Atlas: Test Writing: 29.8%2 Oct 2026
- Terminal-Bench 4.0 (Vals): 2.5%2 Oct 2026
- and 7 more
- MiMo-V2.6-Flash
- SciCode: 51.3%5 Oct 2026
- ProofBench: 63%5 Oct 2026
- LMArena Agent: -0.63 Oct 2026
- LMArena Vision: 126629 Sept 2026
- IOI: 47.7%29 Sept 2026
- BioMysteryBench: 69.3%29 Sept 2026
- and 5 more
- Qwen3.8 27B
- Claude Opus 4.6
- PRBench Finance: 53.3%4 Oct 2026
- MultiChallenge: 37.2%3 Oct 2026
- MultiChallenge: 56%3 Oct 2026
- SWE Atlas: Test Writing: 36.7%2 Oct 2026
- MultiNRC: 57.1%2 Oct 2026
- PRBench Finance: 53.3%2 Oct 2026
- and 5 more
- GLM 5.3 Flash
- CursorBench: 36.8%5 Oct 2026
- FrontierSWE: 18.1%5 Oct 2026
- CursorBench: 26.9%5 Oct 2026
- CursorBench: 31.1%5 Oct 2026
- ARC-AGI-1: 91%2 Oct 2026
- ARC-AGI-2: 65.8%2 Oct 2026
- and 4 more
- Solar Mini4
- AA Intelligence Index: 24.12 Oct 2026
- Humanity's Last Exam (AA): 25.8%2 Oct 2026
- AA-LCR: 83.3%2 Oct 2026
- SciCode (AA): 47.6%2 Oct 2026
- AA-Omniscience: -10.82 Oct 2026
- CritPt: 1.4%2 Oct 2026
- and 4 more
- MiMo-V2.6-Pro
- SciCode: 60.9%5 Oct 2026
- APEX-Agents: 59.5%5 Oct 2026
- ProofBench: 70%5 Oct 2026
- ALE-Bench: 1158.15 Oct 2026
- LMArena Vision: 126529 Sept 2026
- MedCode: 45%29 Sept 2026
- and 3 more
- Muse Spark 1.1
- APEX-Agents: 31.8%5 Oct 2026
- SciCode: 58.8%5 Oct 2026
- PRBench Finance: 55%4 Oct 2026
- MultiChallenge: 75.3%3 Oct 2026
- PRBench Finance: 55%2 Oct 2026
- MultiNRC: 65.6%2 Oct 2026
- and 3 more
- Gemini 3 Pro
- PRBench Finance: 39.2%4 Oct 2026
- AudioMC: 54.7%4 Oct 2026
- MultiChallenge: 65.7%3 Oct 2026
- PRBench Finance: 39.2%2 Oct 2026
- MultiNRC: 59%2 Oct 2026
- MCP Atlas: 70.3%28 Sept 2026
- and 3 more
- Claude Opus 4.5
- PRBench Finance: 46.2%4 Oct 2026
- MultiChallenge: 59%3 Oct 2026
- PRBench Finance: 46.2%2 Oct 2026
- MultiNRC: 48.6%2 Oct 2026
- MultiNRC: 41.2%2 Oct 2026
- TutorBench: 51.2%28 Sept 2026
- and 3 more
- Claude Sonnet 4.5
- PRBench Finance: 43.8%4 Oct 2026
- MultiChallenge: 55.3%3 Oct 2026
- PRBench Finance: 43.8%2 Oct 2026
- MultiNRC: 28.2%2 Oct 2026
- MultiNRC: 35.8%2 Oct 2026
- TutorBench: 45.7%28 Sept 2026
- and 3 more
- GPT-6.1 Sol
- LMArena Text: 14833 Oct 2026
- LMArena Hard Prompts: 15093 Oct 2026
- LMArena Coding: 15423 Oct 2026
- LMArena Vision: 12853 Oct 2026
- LMArena Agent: 11.23 Oct 2026
- LMArena Creative Writing: 14603 Oct 2026
- and 3 more
- Claude Fable 5
- GSO-Bench: 78.4%5 Oct 2026
- PRBench Finance: 53.9%4 Oct 2026
- PRBench Finance: 53.9%2 Oct 2026
- SWE Atlas: Test Writing: 55.6%2 Oct 2026
- Tax Agent Bench: 69.8%29 Sept 2026
- LMArena Vision: 132629 Sept 2026
- and 2 more
- Gemini 3.8 Flash
- WeirdML: 84.8%5 Oct 2026
- Remote Labor Index: 5.8%5 Oct 2026
- AudioMC: 60.4%4 Oct 2026
- PRBench Finance: 49.5%4 Oct 2026
- PRBench Finance: 49.5%2 Oct 2026
- SWE Atlas: Test Writing: 53.7%2 Oct 2026
- and 2 more
- Muse Spark
- PRBench Finance: 52.4%4 Oct 2026
- MultiChallenge: 75.5%3 Oct 2026
- PRBench Finance: 52.4%2 Oct 2026
- MultiNRC: 59.1%2 Oct 2026
- SWE Atlas: Test Writing: 31.1%2 Oct 2026
- SWE Atlas: Codebase QnA: 24.2%29 Sept 2026
- and 2 more
- o3
- PRBench Finance: 47.7%4 Oct 2026
- MultiChallenge: 56.6%3 Oct 2026
- PRBench Finance: 47.7%2 Oct 2026
- MultiNRC: 45.5%2 Oct 2026
- MultiNRC: 44.5%2 Oct 2026
- SciPredict: 17.9%28 Sept 2026
- and 2 more
- Gemini 2.5 Pro
- PRBench Finance: 38.9%4 Oct 2026
- AudioMC: 46.9%4 Oct 2026
- MultiChallenge: 53.6%3 Oct 2026
- PRBench Finance: 38.9%2 Oct 2026
- MultiNRC: 45.1%2 Oct 2026
- TutorBench: 55.7%28 Sept 2026
- and 2 more