Nemotron 3.5 Super VL 120B A12B added (15 Oct 2026)

BenchLeader refreshed 1570 models from 23 of 23 data feeds (benchmark publishers, aggregators and price lists). 1 new model, 8 price changes, 313 new or updated benchmark results across 23 data feeds.

New models

Nemotron 3.5 Super VL 120B A12B (NVIDIA) is now listed; no independent results yet. (details)

Price changes

Nemotron 3 Super 120B A12b input price moved from $0.20/M to $0.30/M (50% up). (details)
Nemotron 3 Super 120B A12b output price moved from $0.80/M to $0.90/M (12% up). (details)
GPT-5.6 Sol Pro (max) input price moved from $2.00/M to $4.00/M (100% up). (details)
GPT-5.6 Sol Pro (max) output price moved from $10.00/M to $20.00/M (100% up). (details)
GPT-5.6 Sol Pro (xhigh) input price moved from $2.00/M to $4.00/M (100% up). (details)
GPT-5.6 Sol Pro (xhigh) output price moved from $10.00/M to $20.00/M (100% up). (details)
GPT-5.6 Sol Pro input price moved from $2.00/M to $4.00/M (100% up). (details)
GPT-5.6 Sol Pro output price moved from $10.00/M to $20.00/M (100% up). (details)

New benchmark results

Chess Puzzles: 13 new results, including Claude Opus 5.5, Gemini 4 Argon, Claude Sonnet 5.5… (board)
Arena Search: 9 new results, including GPT-6 Astra, Claude Fable 5.1, GPT-6.1 Sol… (board)
SciCode: 7 new results, including Muse Wonder, Claude Haiku 5.5, Claude Haiku 5.5… (board)
CursorBench: 6 new results, including Gemini 4 Argon, Claude Haiku 5.5, Claude Haiku 5.5… (board)
ARC-AGI-1: 5 new results, including Claude Haiku 5.5, Claude Haiku 5.5, Claude Haiku 5.5… (board)
GDP.pdf: 4 new results, including Gemini 4 Argon, GPT-6.1 Sol, Muse Wonder… (board)
ALE-Bench: 4 new results, including GPT-6.1 Sol, Muse Spark 1.3, MiMo-V2.6-Flash… (board)
ARC-AGI-2: 4 new results, including Claude Haiku 5.5, Claude Haiku 5.5, Claude Haiku 5.5… (board)
FrontierCode: 3 new results, including Claude Haiku 5.5, DeepSeek V4.1 Flash, Mistral Large 4 (board)
OTIS Mock AIME: 2 new results, including Gemini 4 Argon, Muse Wonder (board)
Mystery Game Puzzles: 2 new results, including Gemini 4 Argon, Mistral Large 4 (board)
Analyst Agent (AA): 2 new results, including Gemini 4 Argon, GLM 5.3 (board)

Score revisions

Claude Opus 5.5 (max) on Arena Agent: 17.9 → 17. (details)
Gemini 4 Argon (high) on Arena Agent: 12.4 → 13.6. (details)
Claude Fable 5.1 (max) on Chess Puzzles: 47% → 48%. (details)
Claude Fable 5.1 (max) on Arena Agent: 11.7 → 14.3. (details)
GPT-6 Astra (max) on Chess Puzzles: 72% → 74%. (details)
GPT-6 Astra (max) on Arena WebDev: 1786 → 1780. (details)
GPT-6 Astra (max) on Arena Agent: 12.3 → 13.7. (details)
GPT-6.1 Sol (max) on Chess Puzzles: 61% → 68%. (details)
GPT-6.1 Sol (max) on Arena Coding: 1545 → 1538. (details)
GPT-6.1 Sol (max) on Arena Agent: 13.8 → 15.1. (details)
GPT-6.1 Sol (max) on Arena Multi-turn: 1487 → 1479. (details)
Claude Fable 5 (max) on Chess Puzzles: 41% → 47%. (details)
Claude Opus 5 (max) on Chess Puzzles: 42% → 43%. (details)
Claude Opus 5 (max) on Arena Agent: 8.2 → 9.2. (details)
Claude Sonnet 5.5 (max) on Arena Agent: 10.8 → 13. (details)
GPT-5.6 Sol (max) on Chess Puzzles: 55% → 59%. (details)
Muse Spark 1.3 (max) on Chess Puzzles: 38% → 37%. (details)
GPT-5.6 Sol (xhigh) on SciCode: 56% → 57.1%. (details)
Claude Sonnet 5.5 (xhigh) on Arena Coding: 1532 → 1527. (details)
Claude Sonnet 5.5 (xhigh) on Arena Instruction Following: 1489 → 1483. (details)
…and 203 more.

Speed changes

Solar Pro 4 time to first token changed from 2.14 s to 0.97 s. (details)
GPT-4 Turbo time to first token changed from 1.15 s to 3.04 s. (details)
GPT-6 Astra Pro time to first token changed from 11.03 s to 24.38 s.

Today's top five

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.