Muse Wonder added, Muse Wonder Full added (13 Oct 2026)

BenchLeader refreshed 1562 models from 23 of 23 data feeds (benchmark publishers, aggregators and price lists). 2 new models, 3 top-10 rank changes, 86 price changes, 212 new or updated benchmark results across 23 data feeds.

New models

Muse Wonder (max) (Meta) appeared with 5 benchmark results, entering the BenchLeader Index at #31. (details)
Muse Wonder Full (max) (Meta) appeared with 4 benchmark results. (details)

Movement in the top 10

Claude Opus 5.5 (high) moved from #4 to #2 in the BenchLeader Index. (details)
Claude Fable 5.1 (max) moved from #2 to #3 in the BenchLeader Index. (details)
GPT-6 Astra (max) moved from #3 to #4 in the BenchLeader Index. (details)

Price changes

GLM 5.3 Flash cost per task moved from 0.3 to 0.3 (12% up). (details)
DeepSeek V4 Flash (max) input price moved from $0.13/M to $0.11/M (15% down). (details)
DeepSeek V4 Flash (max) output price moved from $0.28/M to $0.23/M (18% down). (details)
DeepSeek V4 Flash (max) cost per task moved from 0.1 to 0.1 (17% down). (details)
DeepSeek V4 Flash (high) input price moved from $0.13/M to $0.11/M (15% down). (details)
DeepSeek V4 Flash (high) output price moved from $0.28/M to $0.23/M (18% down). (details)
DeepSeek V4 Flash (high) cost per task moved from 0.1 to 0.1 (17% down). (details)
Ling 3.0 Flash VL input price moved from $0.080/M to $0.075/M (6% down). (details)
GLM 5.3 Flash (max) cost per task moved from 0.3 to 0.3 (12% up). (details)
Qwen3 Max (max) input price moved from $1.20/M to $0.36/M (70% down). (details)
Qwen3 Max (max) output price moved from $6.00/M to $1.43/M (76% down). (details)
Grok 3 input price moved from $4.00/M to $3.00/M (25% down). (details)
Grok 3 output price moved from $20.00/M to $15.00/M (25% down). (details)
Qwen3 VL 235B A22B Instruct (thinking) input price moved from $0.40/M to $0.22/M (45% down). (details)
Qwen3 VL 235B A22B Instruct (thinking) output price moved from $1.60/M to $1.39/M (13% down). (details)
Nova 2.0 Pro Preview (medium) input price moved from $1.25/M to $1.38/M (10% up). (details)
Nova 2.0 Pro Preview (medium) output price moved from $10.00/M to $11.00/M (10% up). (details)
Gemma 4 26B A4B (thinking) output price moved from $0.37/M to $0.40/M (8% up). (details)
Qwen3-Next 80B A3B (thinking) output price moved from $1.20/M to $1.10/M (8% down). (details)
Nova 2.0 Pro Preview (low) input price moved from $1.25/M to $1.38/M (10% up). (details)
Nova 2.0 Pro Preview (low) output price moved from $10.00/M to $11.00/M (10% up). (details)
DeepSeek V4 Pro (no reasoning) cost per task moved from 0.5 to 0.4 (22% down). (details)
Nova 2.0 Lite (thinking) input price moved from $0.30/M to $0.33/M (10% up). (details)
Nova 2.0 Lite (thinking) output price moved from $2.50/M to $2.75/M (10% up). (details)
Qwen3 VL 235B A22B Instruct input price moved from $0.40/M to $0.22/M (45% down). (details)
Qwen3 VL 235B A22B Instruct output price moved from $1.60/M to $1.39/M (13% down). (details)
Nova 2.0 Lite (medium) input price moved from $0.30/M to $0.33/M (10% up). (details)
Nova 2.0 Lite (medium) output price moved from $2.50/M to $2.75/M (10% up). (details)
Qwen3-Next 80B A3B output price moved from $1.20/M to $1.10/M (8% down). (details)
Qwen3 Coder 480B input price moved from $1.50/M to $0.30/M (80% down). (details)
…and 56 more.

New benchmark results

Terminal-Bench-Science 0.1 (AA): 31 new results, including Claude Opus 5.5, Claude Opus 5.5, Claude Fable 5.1… (board)
SWE Atlas: Codebase QnA: 22 new results, including Claude Fable 5.1, GPT-6 Astra, Claude Opus 5… (board)
FrontierCode 1.1 (AA): 21 new results, including Claude Opus 5.5, Claude Fable 5.1, GPT-6 Astra… (board)
AA-3D: 20 new results, including Claude Opus 5.5, Claude Fable 5.1, GPT-6 Astra… (board)
DeepSec-Bench (AA): 18 new results, including Claude Opus 5.5, Claude Fable 5.1, GPT-6 Astra… (board)
AA Cyber Index: 17 new results, including Claude Opus 5.5, Claude Fable 5.1, GPT-6 Astra… (board)
CWE-Bench (AA): 17 new results, including Claude Opus 5.5, Claude Fable 5.1, GPT-6 Astra… (board)
CyberGym end-to-end (AA): 17 new results, including Claude Opus 5.5, Claude Fable 5.1, GPT-6 Astra… (board)
AudioMC (text output): 17 new results, including Inkling Small, Gemini 2.5 Flash, Gemini 2.5 Flash… (board)
Global-MMLU-Lite (AA): 11 new results, including Gemini 3.1 Pro, Gemini 3 Pro, Gemini 3 Flash… (board)
AutomationBench: 2 new results, including Claude Opus 5.5, Claude Sonnet 5.5 (board)
GDP.pdf (AA): 2 new results, including Claude Opus 5.5, Claude Sonnet 5.5 (board)

Score revisions

GPT-6 Astra (max) on GDPval-AA v2.1: 52.1% → 53.7%. (details)
GPT-6 Astra (high) on GDPval-AA v2.1: 49.2% → 51.2%. (details)
GPT-6.1 Sol (max) on GDPval-AA v2.1: 53.8% → 54.6%. (details)
GPT-6 Astra (xhigh) on GDPval-AA v2.1: 50.8% → 52.7%. (details)
GPT-6.1 Sol (high) on GDPval-AA v2.1: 49.3% → 50.5%. (details)
GPT-6 Astra (medium) on GDPval-AA v2.1: 48.4% → 49.3%. (details)
GPT-6.1 Sol (medium) on GDPval-AA v2.1: 46.7% → 47.4%. (details)
GPT-6.1 Sol (xhigh) on GDPval-AA v2.1: 50.5% → 51.9%. (details)
GPT-6 Astra (low) on GDPval-AA v2.1: 43.3% → 44.5%. (details)
GPT-6.1 Sol (low) on GDPval-AA v2.1: 39.9% → 40.8%. (details)
Claude Opus 4.5 on BALROG: 43.5% → 43%. (details)

Speed changes

Claude Opus 5.5 time to first answer changed from 36.37 s to 49.71 s. (details)
Claude Fable 5.1 time to first token changed from 2.46 s to 3.56 s. (details)
GPT-6 Astra output speed changed from 47 tok/s to 66 tok/s. (details)
GPT-6 Astra time to first answer changed from 83.11 s to 114.18 s. (details)
Claude Fable 5.1 time to first answer changed from 85.17 s to 118.26 s. (details)
Claude Fable 5.1 response time changed from 93.46 s to 126.37 s. (details)
Claude Opus 5 time to first answer changed from 58.93 s to 86.18 s. (details)
Claude Opus 5 response time changed from 68.64 s to 95.17 s. (details)
Muse Spark 1.3 thinking time changed from 11.41 s to 15.72 s. (details)
GPT-5.6 Sol time to first answer changed from 85.27 s to 128.56 s. (details)
GPT-5.6 Sol response time changed from 92.04 s to 134.34 s. (details)
GPT-5.5 time to first answer changed from 38.96 s to 70.28 s. (details)
GPT-5.5 response time changed from 44.67 s to 75.42 s. (details)
GPT-6 Sol output speed changed from 86 tok/s to 122 tok/s. (details)
GPT-6 Sol time to first token changed from 1.38 s to 2.66 s. (details)
…and 160 more.

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.