Daily digest, 28 Sept 2026
2 new models, 2 top-10 rank changes, 12 price changes, 93 new or updated benchmark results across 22 sources.
BenchLeader refreshed 1525 models from 22 of 22 sources. 2 new models, 2 top-10 rank changes, 12 price changes, 93 new or updated benchmark results across 22 sources.
New models
- Daybreak Blue (OpenAI) is now listed at $8.00/M blended; no independent results yet. (details)
- Daybreak Red (OpenAI) is now listed at $28.13/M blended; no independent results yet. (details)
Movement in the top 10
- Claude Opus 5.5 (high) moved from #8 to #7 in the BenchLeader Index. (details)
- GPT-6 Astra (xhigh) moved from #7 to #8 in the BenchLeader Index. (details)
Price changes
- Qwen3.7 Plus input price moved from $0.50/M to $0.40/M (20% down). (details)
- Qwen3.7 Plus output price moved from $3.00/M to $1.60/M (47% down). (details)
- Qwen3.7 Plus cached input price moved from $0.05/M to $0.04/M (20% down). (details)
- Qwen3.5 Flash input price moved from $0.03/M to $0.10/M (245% up). (details)
- Qwen3.5 Flash output price moved from $0.29/M to $0.40/M (39% up). (details)
- Qwen3.7 Flash (no reasoning) output price moved from $0.12/M to $0.13/M (10% up). (details)
- Qwen3.7 Flash output price moved from $0.12/M to $0.13/M (10% up). (details)
- Qwen3.7 Plus (no reasoning) input price moved from $0.50/M to $0.40/M (20% down). (details)
- Qwen3.7 Plus (no reasoning) output price moved from $3.00/M to $1.60/M (47% down). (details)
- Qwen3.7 Plus (no reasoning) cached input price moved from $0.05/M to $0.04/M (20% down). (details)
- Qwen3.5 Flash (no reasoning) input price moved from $0.03/M to $0.10/M (245% up). (details)
- Qwen3.5 Flash (no reasoning) output price moved from $0.29/M to $0.40/M (39% up). (details)
New benchmark results
- MCP Atlas: 32 new results, including Claude Opus 5, Claude Fable 5, GPT-5.5… (board)
- TutorBench: 26 new results, including Muse Spark, GPT-5.4 Pro, Gemini 3.1 Pro… (board)
- AudioMC (text output): 17 new results, including Inkling Small, Gemini 2.5 Flash, Gemini 2.5 Flash… (board)
- SciPredict: 15 new results, including Gemini 3 Pro, o3, Claude Opus 4.5… (board)
- LMArena Agent: 1 new result, including Claude Opus 5.5 (board)
Score revisions
- GPT-6 Astra (max) on LMArena Agent: 10.8 → 10.3. (details)
- Grok 4.7 (xhigh) on LMArena Agent: 3.8 → 4.4. (details)
Speed changes
- Kimi K2.6 output speed changed from 80 tok/s to 38 tok/s. (details)
- Kimi K2.6 time to first token changed from 2.85 s to 1.19 s. (details)
- GLM 4.7 time to first token changed from 5.10 s to 1.42 s. (details)
- Mistral Medium 3 time to first token changed from 1.06 s to 0.62 s. (details)
- GPT-4 output speed changed from 54 tok/s to 25 tok/s. (details)
- Gemma 3 27B time to first token changed from 0.86 s to 1.20 s. (details)
- Nemotron 3 Super 120B A12b output speed changed from 23 tok/s to 39 tok/s. (details)
- Qwen3 8B output speed changed from 14 tok/s to 50 tok/s. (details)
- GPT-5.2 Pro output speed changed from 2 tok/s to 5 tok/s. (details)
- ERNIE 4.5 VL 424B A47B output speed changed from 16 tok/s to 10 tok/s. (details)
Today's top five
- Claude Opus 5.5 — 72.0
- Claude Fable 5.1 — 70.8
- GPT-6 Astra — 70.2
- Claude Opus 5.5 — 70.2
- Gemini 4 Argon — 70.0
This digest is generated automatically from the day's data changes. Every line links to the page where you can check the numbers and their source.