agnes-3-0-flash: new AI model, 7 Sep–13 Sep
1 new model incl. agnes-3-0-flash, GPT-5 mini up 8.3, 37 price cuts, 3 new benchmark results: what moved on the BenchLeader Index this week.
The week of 7 Sep 2026 to 13 Sep 2026, from 6 daily updates.
Three things that happened
- Biggest riser: GPT-5 mini, up 8.3 points to 58.8 on the BenchLeader Index as new results landed.
- 1 new model arrived: agnes-3-0-flash.
- Busiest benchmark: AA Intelligence Index v4.3.2 with 1 new result; 3 new results were published in all.
This week BenchLeader added sources, which listed their back catalogues on 9 Sep 2026, 10 Sep 2026, 12 Sep 2026 and 13 Sep 2026. Those models and results are left out of the counts above: they are new to this site, not new to the world.
Index movers
- GPT-5 mini+8.3
- MiniMax M2.7+7.2
- Magistral Medium-6.8
- MiniMax M2.5+6.7
- gpt-oss-120b+6.6
- GLM 5.1+6.5
- MiniMax M3+5.8
- Qwen3 235B A22B-5.8
| Model | Provider | 7 Sep 2026 | 13 Sep 2026 | Change | |---|---|---:|---:|---:| | GPT-5 mini | OpenAI | 50.5 | 58.8 | +8.3 | | MiniMax M2.7 | MiniMax | 49.4 | 56.6 | +7.2 | | Magistral Medium | Mistral AI | 46.5 | 39.7 | -6.8 | | MiniMax M2.5 | MiniMax | 46.7 | 53.4 | +6.7 | | gpt-oss-120b | OpenAI | 42.5 | 49.1 | +6.6 | | GLM 5.1 | Zhipu AI | 53.6 | 60.1 | +6.5 | | MiniMax M3 | MiniMax | 52.3 | 58.1 | +5.8 | | Qwen3 235B A22B | Alibaba | 51.7 | 45.9 | -5.8 | | Qwen3 30B A3B | Alibaba | 48.6 | 42.9 | -5.7 | | Grok 4 Fast | SpaceXAI | 46.7 | 52.2 | +5.5 |
New models
- Agnes 3.0 Flash (Sapiens AI) appeared with 3 benchmark results. (details)
Price changes
- Muse Spark 1.2 (xhigh) output price moved from $4.25/M to $0.20/M (95% down).
- Muse Spark 1.3 (max) output price moved from $4.25/M to $0.20/M (95% down).
- MiMo-V2.5 output price moved from $2.00/M to $0.28/M (86% down).
- Qwen3.5 Flash (no reasoning) output price moved from $1.72/M to $0.26/M (85% down).
- GLM 5.2 (max) output price moved from $3.85/M to $0.88/M (77% down).
- Qwen3 32B (no reasoning) output price moved from $1.15/M to $0.28/M (76% down).
- Mistral Medium 3 input price moved from $1.50/M to $0.40/M (73% down).
- Gemma 4 26B A4B (minimal) input price moved from $0.13/M to $0.04/M (68% down).
- MiMo-V2.5-Pro output price moved from $3.00/M to $1.04/M (65% down).
- LongCat 2.0 input price moved from $0.75/M to $0.30/M (60% down).
- Qwen3.7 Plus (no reasoning) output price moved from $3.00/M to $1.28/M (57% down).
- GPT-5.6 Sol (max) input price moved from $4.40/M to $2.00/M (55% down).
- GPT-4o (high) input price moved from $5.00/M to $2.50/M (50% down).
- Qwen3.5 Plus 2026-04-20 (thinking) output price moved from $3.44/M to $1.80/M (48% down).
- Qwen3.8 27B (xhigh) input price moved from $0.80/M to $0.42/M (48% down).
Top five as of 13 Nov 2026
- Claude Fable 5.2 — 72.7
- Gemini 4 Carbon — 69.5
- Claude Opus 5.5 — 68.4
- GPT-6 Astra — 67.1
- Claude Fable 5.1 — 66.6
More: New AI model releases this week, and the full LLM leaderboard 2026.
Generated from the week's daily updates; every line links to the page where the numbers and their sources can be checked.
The week in dated, citable sentences: state of the leaderboard, week 37.