musespark-1-3 added, LFM2.5-24B-A2B added (14 Oct 2026)

BenchLeader refreshed 1564 models from 23 of 23 data feeds (benchmark publishers, aggregators and price lists). 2 new models, 4 top-10 rank changes, 3 price changes, 122 new or updated benchmark results across 23 data feeds.

New models

Musespark 1.3 (Meta) appeared with 1 benchmark result. (details)
Lfm 2.5 24B A2b (Liquid AI) is now listed at $0.21/M blended; no independent results yet. (details)

Movement in the top 10

Gemini 4 Argon (high) moved from #5 to #2 in the BenchLeader Index. (details)
GPT-6 Astra (high) moved from #6 to #5 in the BenchLeader Index. (details)
Claude Fable 5.1 (xhigh) moved from #7 to #6 in the BenchLeader Index. (details)
Claude Opus 5.5 (high) moved from #2 to #7 in the BenchLeader Index. (details)

Price changes

Muse Glimmer (high) input price moved from $0.32/M to $0.30/M (6% down). (details)
Muse Glimmer (high) output price moved from $1.35/M to $1.20/M (11% down). (details)
Muse Glimmer (high) cost per task moved from 0.1 to 0.1 (6% down). (details)

New benchmark results

PRBench Legal: 39 new results, including GPT-6 Astra, Claude Fable 5, GPT-5.6 Sol… (board)
AudioMC (audio output): 15 new results, including Qwen3 Omni 30B A3B Instruct, Gemini 2.5 Flash Native Audio 12 2025, Gemini 2.5 Flash Native Audio 12 2025… (board)
HiL-Bench: 8 new results, including GPT-6 Astra, GPT-6.1 Sol, Kimi K3… (board)
Analyst Agent (AA): 8 new results, including Muse Spark 1.3, MiMo-V2.6-Pro, Gemini 3.8 Flash… (board)
ARC-AGI-3: 4 new results, including DeepSeek V4.1 Flash, DeepSeek V4.1 Flash, DeepSeek V4.1 Flash… (board)
FACTS Search: 3 new results, including GPT-6.1 Sol, Gemini 4 Argon, Claude Haiku 5.5 (board)
Arena Agent: 2 new results, including Claude Opus 5.5, Nemotron 3 Ultra (board)
GPQA Diamond: 2 new results, including Gemini 4 Argon, Mistral Large 4 (board)
SimpleQA Verified: 2 new results, including Gemini 4 Argon, DeepSeek V4.1 Flash (board)
Harvey LAB-AA v1.1: 2 new results, including Muse Wonder, Muse Wonder Full (board)
Terminal-Bench-Science 0.1 (AA): 2 new results, including Grok 4.7, K2 Horizon 375B A23B (board)
EBR-bench: 1 new result, including Muse Wonder (board)

Score revisions

Gemini 4 Argon (high) on Arena Agent: 9.3 → 12.4. (details)
Claude Fable 5.1 (max) on Arena Agent: 12.7 → 11.7. (details)
GPT-6 Astra (max) on Arena Agent: 13.1 → 12.3. (details)
GPT-6.1 Sol (max) on Arena Agent: 11.7 → 13.8. (details)
Muse Spark 1.3 (max) on Arena Agent: 4 → 3.3. (details)
Claude Sonnet 5.5 (max) on Arena Agent: 12 → 10.8. (details)
GPT-5.6 Sol (xhigh) on Arena Agent: 6.2 → 7.6. (details)
GPT-6 Sol (max) on Arena Agent: 9.9 → 12.6. (details)
Kimi K3 (max) on Arena Agent: 3.8 → 3. (details)
Gemini 3.8 Flash (high) on Arena Agent: 2.8 → 5.2. (details)
Step 5 Preview on Arena Agent: 0 → -1. (details)
Gemini 3.7 Flash (high) on Arena Agent: -2.1 → -0.7. (details)
Muse Spark 1.2 (xhigh) on Arena Agent: -3.7 → -6.5. (details)
GLM 5.3 Flash on Arena Agent: -1 → -1.5. (details)
Gemini 3.1 Pro on Arena Agent: -7.8 → -2.8. (details)
DeepSeek V4.1 Flash (max) on Arena Agent: 3.8 → 3.3. (details)
Claude Opus 4.8 (high) on Arena Agent: 7.1 → 7.8. (details)
Gemini 3.6 Flash (high) on Arena Agent: -7.8 → -4.4. (details)
Grok 4.7 (xhigh) on Arena Agent: 2.8 → 3.6. (details)
Claude Sonnet 5 (high) on Arena Agent: 4.9 → 4.3. (details)
…and 12 more.

Speed changes

Claude Fable 5.1 time to first answer changed from 1.95 s to 3.05 s. (details)
Qwen3.8 2.4T A95B time to first token changed from 3.67 s to 1.21 s. (details)
Step 5 Preview time to first token changed from 1.82 s to 3.09 s. (details)
Claude Opus 4.7 time to first answer changed from 21.49 s to 37.14 s. (details)
Claude Opus 4.7 response time changed from 31.54 s to 46.71 s. (details)
GLM 5.3 Flash time to first token changed from 0.39 s to 0.79 s. (details)
GPT-5.6 Terra time to first answer changed from 7.36 s to 10.91 s. (details)
Claude Sonnet 5 time to first answer changed from 12.50 s to 17.81 s. (details)
Grok 4.5 time to first token changed from 4.17 s to 1.91 s. (details)
GPT-5 time to first token changed from 3.19 s to 4.52 s. (details)
GLM 5.1 time to first token changed from 9.35 s to 1.98 s. (details)
Gemma 4 26B A4B output speed changed from 82 tok/s to 49 tok/s. (details)
Qwen3.8 27B time to first token changed from 1.72 s to 0.69 s. (details)
GLM 5V Turbo output speed changed from 26 tok/s to 37 tok/s. (details)
Ling 3.0 Flash VL output speed changed from 47 tok/s to 145 tok/s. (details)
…and 56 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.