Daily digest, 15 Sept 2026

BenchLeader refreshed 1360 models from 20 of 20 sources. 1 new model, 2 top-10 rank changes, 5 price changes, 13 new or updated benchmark results across 20 sources.

New models

• Mercury 2.5 (high) (Inception) appeared with 3 benchmark results. (details)

Movement in the top 10

• Claude Opus 5 moved from #9 to #8 in the BenchLeader Index. (details)
• GPT-6 Astra (max) moved from #8 to #9 in the BenchLeader Index. (details)

Price changes

• Qwen3.8 27B cached input price moved from $0.04/M to $0.05/M (25% up). (details)
• Qwen3.8 27B (low) cached input price moved from $0.04/M to $0.05/M (25% up). (details)
• Qwen3.8 27B (medium) cached input price moved from $0.04/M to $0.05/M (25% up). (details)
• Qwen3.8 27B (no reasoning) cached input price moved from $0.04/M to $0.05/M (25% up). (details)
• Qwen3.8 27B (xhigh) cached input price moved from $0.04/M to $0.05/M (25% up). (details)

New benchmark results

• SciCode (AA): 2 new results, including GPT-5.6 Sol, GPT-5.6 Terra (board)
• LMArena Agent: 1 new result, including DeepSeek V4.1 Flash (board)

Score revisions

• GPT-6 Astra (max) on LMArena Agent: 12.4 → 11.9. (details)
• Claude Opus 5 (high) on LMArena Agent: 11.1 → 10.5. (details)
• Grok 4.6 (xhigh) on LMArena Agent: 3.1 → 2.6. (details)
• Claude Sonnet 5 (high) on LMArena Agent: 5.9 → 5.4. (details)
• GLM 5.3 Flash on LMArena Agent: 1.9 → 1.3. (details)
• Kimi K2.6 on AA Intelligence Index: 31.3 → 27.5. (details)
• GLM 5.3 (max) on LMArena Agent: 2.6 → 3.1. (details)
• Inkling Small on LMArena Agent: -7.1 → -7.8. (details)
• Gemini 3.5 Flash Lite on LMArena Agent: -13.7 → -14.2. (details)
• Ling 3.0 Flash on AA Intelligence Index: 24.9 → 20.6. (details)

Speed changes

• Claude Sonnet 5 time to first answer changed from 4.78 s to 2.76 s. (details)
• Mistral Medium 3.5 time to first token changed from 0.47 s to 0.66 s. (details)
• Mercury 2 time to first token changed from 0.86 s to 0.53 s. (details)
• Qwen2.5 72B time to first token changed from 1.02 s to 0.66 s. (details)
• Nemotron Nano 12B v2 VL output speed changed from 109 tok/s to 148 tok/s. (details)
• Gemma 3 4B time to first token changed from 1.07 s to 1.59 s. (details)
• Palmyra X5 output speed changed from 48 tok/s to 28 tok/s. (details)
• Devstral 2 output speed changed from 19 tok/s to 31 tok/s. (details)
• Lyria 3 Pro output speed changed from 4 tok/s to 2 tok/s. (details)
• Lyria 3 Pro time to first token changed from 2.12 s to 3.63 s. (details)

Today's top five

1. Claude Fable 5.1 — 70.5
2. Claude Fable 5.1 — 70.4
3. GPT-6 Astra — 70.2
4. Claude Fable 5 — 70.1
5. Claude Fable 5.1 — 69.9

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.