Muse Wonder Full up to #3, GLM 4.5V price cut (20 Oct 2026)

BenchLeader refreshed 1602 models from 24 of 24 data feeds (benchmark publishers, aggregators and price lists). 8 top-10 rank changes, 85 price changes, 194 new or updated benchmark results across 24 data feeds.

Movement in the top 10

Muse Wonder Full (max) moved from #15 to #3 in the BenchLeader Index. (details)
GPT-6 Astra (max) moved from #3 to #4 in the BenchLeader Index. (details)
Claude Fable 5.1 (max) moved from #4 to #5 in the BenchLeader Index. (details)
GPT-6.1 Sol (max) moved from #5 to #7 in the BenchLeader Index. (details)
GPT-6 Astra (high) moved from #7 to #8 in the BenchLeader Index. (details)
Claude Sonnet 5.5 (max) moved from #8 to #9 in the BenchLeader Index. (details)
Claude Fable 5.1 (xhigh) moved from #9 to #10 in the BenchLeader Index. (details)
Claude Fable 5 (max) moved from #10 to #11 in the BenchLeader Index. (details)

Price changes

GLM 5.3 (max) cached input price moved from $0.26/M to $0.30/M (15% up). (details)
GLM 5.3 Flash cached input price moved from $0.030/M to $0.034/M (15% up). (details)
GLM 5.2 (max) cached input price moved from $0.26/M to $0.30/M (15% up). (details)
Qwen3 7 (max) cached input price moved from $0.50/M to $0.33/M (34% down). (details)
GLM-5 (thinking) cached input price moved from $0.20/M to $0.15/M (25% down). (details)
Qwen3.6 Plus cached input price moved from $0.050/M to $0.028/M (45% down). (details)
Qwen3.7 Plus cached input price moved from $0.050/M to $0.028/M (45% down). (details)
GLM 5.1 (thinking) cached input price moved from $0.26/M to $0.19/M (25% down). (details)
GLM 5.1 input price moved from $1.40/M to $0.90/M (36% down). (details)
GLM 5.1 output price moved from $4.40/M to $3.59/M (19% down). (details)
GLM 5.1 cached input price moved from $0.26/M to $0.19/M (25% down). (details)
GLM 5V Turbo input price moved from $1.20/M to $0.75/M (38% down). (details)
GLM 5V Turbo output price moved from $4.00/M to $3.29/M (18% down). (details)
GLM 5V Turbo cached input price moved from $0.24/M to $0.18/M (25% down). (details)
GLM 5.3 (low) cached input price moved from $0.26/M to $0.30/M (15% up). (details)
GLM-5 input price moved from $1.00/M to $0.60/M (40% down). (details)
GLM-5 output price moved from $3.20/M to $2.69/M (16% down). (details)
GLM-5 cached input price moved from $0.20/M to $0.15/M (25% down). (details)
GLM 5.3 Flash (max) cached input price moved from $0.030/M to $0.034/M (15% up). (details)
GLM 5V Turbo (thinking) input price moved from $1.20/M to $0.75/M (38% down). (details)
GLM 5V Turbo (thinking) output price moved from $4.00/M to $3.29/M (18% down). (details)
GLM 5V Turbo (thinking) cached input price moved from $0.24/M to $0.18/M (25% down). (details)
GLM-5 (no reasoning) cached input price moved from $0.20/M to $0.15/M (25% down). (details)
GLM 4.7 (thinking) cached input price moved from $0.11/M to $0.090/M (19% down). (details)
GLM 4.7 input price moved from $0.60/M to $0.45/M (25% down). (details)
GLM 4.7 cached input price moved from $0.11/M to $0.090/M (19% down). (details)
GLM 5.2 input price moved from $1.40/M to $1.20/M (15% down). (details)
GLM 5.2 cached input price moved from $0.26/M to $0.30/M (15% up). (details)
GLM 4.5V input price moved from $0.60/M to $0.30/M (50% down). (details)
GLM 4.5V output price moved from $1.80/M to $0.90/M (50% down). (details)
…and 55 more.

New benchmark results

ForecastBench: 14 new results, including Claude Fable 5, Claude Opus 5, GPT-5.6 Sol… (board)
Arena Vision: 8 new results, including Mistral Large 4, Kimi K2.5, Gemini 1.5 Pro… (board)
FrontierCode 1.1 (AA): 4 new results, including DeepSeek V4.1 Flash 1017, DeepSeek V4 Pro, MiniMax M3.1 Flash… (board)
Arena WebDev: 3 new results, including Muse Wonder Full, MiniMax M3.1 Flash, Nemotron 3.5 Super VL 120B A12B (board)
Terminal-Bench-Science 0.1 (AA): 2 new results, including Nemotron 3 Ultra 550B A55B, Nemotron 3.5 Super VL (board)
Harvey LAB-AA v1.1: 1 new result, including Gemini 4 Argon (board)
LiveBench 2.0: 1 new result, including MiniMax M3.1 Flash (board)
LiveBench Long Agentic: 1 new result, including MiniMax M3.1 Flash (board)
LiveBench 2.0 Reasoning: 1 new result, including MiniMax M3.1 Flash (board)
LiveBench 2.0 Coding: 1 new result, including MiniMax M3.1 Flash (board)
LiveBench Browser: 1 new result, including MiniMax M3.1 Flash (board)
LiveBench 2.0 Data Analysis: 1 new result, including MiniMax M3.1 Flash (board)

Score revisions

Claude Opus 5.5 (max) on AA Cyber Index: 2866.4 → 28.7. (details)
Claude Opus 5.5 (max) on Arena Agent: 18.9 → 20.4. (details)
Gemini 4 Argon (high) on Arena Agent: 13.2 → 12.1. (details)
Gemini 4 Argon (high) on Harvey's Legal Agent Benchmark: 19.6% → 27.1%. (details)
Muse Wonder Full (max) on Vals Index: 64.6 → 66.8. (details)
Muse Wonder Full (max) on Harvey's Legal Agent Benchmark: 21.7% → 23.3%. (details)
GPT-6 Astra (max) on AA Cyber Index: 3342 → 33.4. (details)
GPT-6 Astra (max) on Arena Agent: 14.1 → 15.1. (details)
GPT-6 Astra (max) on Vals Index: 63.1 → 65. (details)
GPT-6 Astra (max) on Harvey's Legal Agent Benchmark: 5.4% → 7.1%. (details)
Claude Fable 5.1 (max) on AA Cyber Index: 2502 → 25. (details)
GPT-6.1 Sol (max) on Vals Index: 61.2 → 63.9. (details)
GPT-6.1 Sol (max) on Harvey's Legal Agent Benchmark: 5.4% → 6.3%. (details)
Claude Opus 5 (max) on Arena Agent: 8.7 → 8.1. (details)
Muse Wonder (max) on Arena Agent: 4.9 → 3.8. (details)
Muse Wonder (max) on Vals Index: 61.1 → 64.9. (details)
Muse Wonder (max) on Harvey's Legal Agent Benchmark: 17.1% → 23.8%. (details)
Muse Spark 1.3 (max) on Arena Agent: 3.8 → 3.3. (details)
GPT-5.6 Sol (xhigh) on Arena Agent: 7.1 → 6.6. (details)
GPT-6 Sol (max) on AA Cyber Index: 3682.8 → 36.8. (details)
…and 125 more.

Speed changes

Gemini 4 Argon output speed changed from 39 tok/s to 84 tok/s. (details)
GPT-6 Astra time to first token changed from 2.75 s to 4.00 s. (details)
Claude Sonnet 5.5 time to first token changed from 1.38 s to 1.98 s. (details)
Claude Opus 5 time to first answer changed from 44.14 s to 68.80 s. (details)
Claude Opus 5 response time changed from 53.45 s to 78.09 s. (details)
GPT-6 Sol time to first answer changed from 32.66 s to 50.25 s. (details)
GPT-6 Sol response time changed from 36.94 s to 54.52 s. (details)
Claude Opus 4.6 output speed changed from 22 tok/s to 40 tok/s. (details)
DeepSeek V4 Flash time to first token changed from 0.91 s to 1.27 s. (details)
Step 5 Preview time to first token changed from 0.68 s to 2.07 s. (details)
Claude Sonnet 5 time to first answer changed from 4.14 s to 7.69 s. (details)
MiMo-V2.6-Flash time to first token changed from 2.59 s to 4.32 s. (details)
Kimi K3 output speed changed from 34 tok/s to 50 tok/s. (details)
Claude Sonnet 5.5 time to first answer changed from 1.29 s to 1.80 s. (details)
Grok 4.20 Multi-Agent time to first token changed from 8.76 s to 5.47 s. (details)
…and 69 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.