Kimi K3.1 Pro Together added, Inkling 2 Small added (6 Nov)
Kimi K3.1 Pro Together and Inkling 2 Small join the board. 2 new models, 2 top-10 rank changes, 7 price changes, 52 new benchmark results. Sources linked.
BenchLeader refreshed 1695 models from 27 of 27 data feeds (benchmark publishers, aggregators and price lists). 2 new models, 2 top-10 rank changes, 7 price changes, 52 new or updated benchmark results across 27 data feeds.
New models
- Kimi K3.1 Pro Together (max) (Moonshot AI) appeared with 5 benchmark results, entering the BenchLeader Index at #75. (details)
- Inkling 2 Small (Thinking Machines) appeared with 8 benchmark results, entering the BenchLeader Index at #146. (details)
Movement in the top 10
- GPT-6.1 Sol (max) moved from #6 to #5 in the BenchLeader Index. (details)
- Claude Fable 5.1 (max) moved from #5 to #6 in the BenchLeader Index. (details)
Price changes
- Nemotron 3.5 Super VL input price moved from $0.20/M to $0.18/M (10% down). (details)
- DeepSeek V3.1 (thinking) input price moved from $0.58/M to $0.43/M (26% down). (details)
- DeepSeek V3.1 (thinking) output price moved from $1.68/M to $1.34/M (20% down). (details)
- Nemotron 3 Super 120B A12B output price moved from $2.40/M to $0.90/M (63% down). (details)
- Nemotron 3.5 Super VL (no reasoning) input price moved from $0.20/M to $0.18/M (10% down). (details)
- Nemotron 3.5 Super VL (medium) input price moved from $0.20/M to $0.18/M (10% down). (details)
- Nemotron 3.5 Super VL (high) input price moved from $0.20/M to $0.18/M (10% down). (details)
New benchmark results
- HLE-Diamond (AA): 3 new results, including GLM 5.2, Solar Pro 4, k-exaone-2-0-0803 (board)
- AA-Omniscience v1.1: 3 new results, including GLM 5.2, Solar Pro 4, k-exaone-2-0-0803 (board)
- AA-Omniscience v1.1: accuracy: 3 new results, including GLM 5.2, Solar Pro 4, k-exaone-2-0-0803 (board)
- AA-Omniscience v1.1: non-hallucination: 3 new results, including GLM 5.2, Solar Pro 4, k-exaone-2-0-0803 (board)
- CritPt v2 hard subset (AA): 3 new results, including Solar Pro 4, k-exaone-2-0-0803, K2 Horizon MoVA 36B A4B (board)
- FACTS Search: 2 new results, including Gemini 4 Argon, Gemini 4 Carbon (board)
- MCP Atlas: 2 new results, including Gemini 4 Argon, Claude Opus 5.5 (board)
- AA Intelligence Index v5.0.1: 2 new results, including k-exaone-2-0-0803, North Mini Code (board)
- FrontierCode 2.0 (AA): 2 new results, including k-exaone-2-0-0803, North Mini Code (board)
- Terminal-Bench-Science 0.1 (AA): 2 new results, including k-exaone-2-0-0803, North Mini Code (board)
- Terminal-Bench 4.0 (AA): 1 new result, including k-exaone-2-0-0803 (board)
- AutomationBench: 1 new result, including k-exaone-2-0-0803 (board)
Score revisions
- GPT-6 Astra on FACTS Search: 0.9% → 93.3%. (details)
- Claude Opus 5.5 on FACTS Search: 0.9% → 88.4%. (details)
- GPT-6.1 Sol on FACTS Search: 0.9% → 92.8%. (details)
- Gemini 3.1 Pro on FACTS Search: 0.9% → 90.2%. (details)
- Kimi K3 on FACTS Search: 0.9% → 90.3%. (details)
- Qwen3.8 Max (max) on FACTS Search: 0.9% → 85%. (details)
- GPT-6 Luna (max) on OpenRouter token share: 4.6% → 3.9%. (details)
- Gemma 4 26B A4B on FACTS Search: 0.7% → 66.4%. (details)
- Hy4 on OpenRouter token share: 0.6% → 2%. (details)
- Gemini 3.8 Flash on FACTS Search: 0.8% → 80.4%. (details)
- GPT-6 Luna (xhigh) on OpenRouter token share: 4.6% → 3.9%. (details)
- Gemini 3.5 Flash Lite on FACTS Search: 0.7% → 69%. (details)
- GPT-6 Luna (high) on OpenRouter token share: 4.6% → 3.9%. (details)
- GPT-6 Luna (medium) on OpenRouter token share: 4.6% → 3.9%. (details)
- GPT-6 Luna (low) on OpenRouter token share: 4.6% → 3.9%. (details)
- gpt-oss-120b on FACTS Search: 0.6% → 57.6%. (details)
- GPT-6 Luna (no reasoning) on OpenRouter token share: 4.6% → 3.9%. (details)
- Claude Haiku 5.5 on FACTS Search: 0.8% → 80.3%. (details)
- Claude Sonnet 5.5 on FACTS Search: 0.9% → 86.8%. (details)
- GLM 5.3 on FACTS Search: 0.9% → 87.7%. (details)
- …and 3 more.
Speed changes
- Claude Opus 5.5 time to first answer changed from 627.31 s to 366.70 s. (details)
- Gemini 4 Carbon time to first answer changed from 152.07 s to 67.72 s. (details)
- GPT-6 Astra time to first token changed from 6.79 s to 4.13 s. (details)
- GPT-6 Astra output speed changed from 67 tok/s to 41 tok/s. (details)
- GPT-6 Astra response time changed from 355.71 s to 540.17 s. (details)
- GPT-6.1 Sol time to first answer changed from 321.76 s to 193.87 s. (details)
- GPT-6.1 Sol response time changed from 328.51 s to 455.09 s. (details)
- GPT-6.1 Sol time to first token changed from 8.86 s to 3.87 s. (details)
- Claude Fable 5.1 response time changed from 272.58 s to 439.75 s. (details)
- GPT-6 Astra time to first answer changed from 121.80 s to 53.44 s. (details)
- Claude Opus 5.5 response time changed from 152.11 s to 222.31 s. (details)
- Gemini 4 Argon response time changed from 51.06 s to 115.89 s. (details)
- Claude Opus 5.5 output speed changed from 78 tok/s to 107 tok/s. (details)
- Claude Opus 5 response time changed from 85.94 s to 305.88 s. (details)
- GPT-5.6 Sol response time changed from 90.19 s to 167.17 s. (details)
- …and 570 more.
Today's top five
- Claude Fable 5.2 (max) — 72.7
- Gemini 4 Carbon (high) — 68.9
- Claude Opus 5.5 (max) — 68.4
- GPT-6 Astra (max) — 67.3
- Claude Fable 5.1 (max) — 66.7
More: New AI model releases this week, and the full LLM leaderboard 2026.
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.