Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Muse Spark 1.3

Meta

Strongest in Legal (#1 of 151), weakest in Mathematics (#26 of 140). Above par in 19 of 19 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5.1 and Claude Fable 5 and ahead of GPT-6 Astra and GPT-5.5.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Professional60.2#3/168
Legal67.1#1/151
Cybersecurity57.3#4/9
Finance58.8#9/151
Core abilities63.3#5/204
Instruction following67.7#4/57
General intelligence65.8#6/204
Data analysis60.4#8/57
Language56.6#19/57
Agents64.7#7/268
Knowledge work69.1#6/178
Human preference66.1#7/342
Human preference66.1#7/342
Coding61.1#8/165
Agentic coding62.4#7/157
Code generation57.6#18/77
Reasoning60#11/178
Reasoning63.8#9/140
Mathematics57.2#26/140

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Muse Spark 1.3, left for the other.

Nemotron 3 Ultrawon 12 · lost 0
Llama 4 Scout Instructwon 12 · lost 0
Command Awon 12 · lost 0
Gemini 2.0 Flashwon 12 · lost 0
Grok 3 Miniwon 12 · lost 0
MiniMax M2.5won 12 · lost 0
Grok 4 Fastwon 12 · lost 0
Inkling Smallwon 12 · lost 0
GLM-5won 12 · lost 0
GLM-5.1won 12 · lost 0
Kimi K2.7 Codewon 13 · lost 0
Grok Build 0.1won 13 · lost 0
Qwen3.8 Flash Nextwon 13 · lost 0
DeepSeek V4 Flash Visionwon 13 · lost 0
O3 Miniwon 13 · lost 0
Grok 4won 13 · lost 0
GPT-4.1won 13 · lost 0
Kimi K2won 13 · lost 0
Qwen3 235B A22B Instructwon 13 · lost 0
GLM-4.6won 13 · lost 0
Mistral Medium 3.5won 13 · lost 0
Claude Opus 4won 13 · lost 0
Gemini 3.1 Flash Litewon 13 · lost 0
Gemma 4 31Bwon 13 · lost 0
Mimo V2.5 Prowon 13 · lost 0
GPT-4o Miniwon 14 · lost 0
GPT-4.1 Nanowon 14 · lost 0
GPT OSS 120Bwon 14 · lost 0
GPT-4.1 Miniwon 14 · lost 0
GPT-5 Miniwon 14 · lost 0
Gemini 2.5 Flashwon 14 · lost 0
Mistral Large 3won 14 · lost 0
Claude Haiku 4.5won 14 · lost 0
DeepSeek V3.2won 14 · lost 0
Claude Sonnet 4.5won 14 · lost 0
DeepSeek V3won 15 · lost 0
Claude Sonnet 4won 15 · lost 0
Qwen3.6 27Bwon 16 · lost 0
Claude Opus 4.5won 17 · lost 0
GPT-5.4 Nanowon 18 · lost 0
Qwen3.8 27Bwon 18 · lost 0
MiniMax M3won 18 · lost 0
GPT-6 Lunawon 18 · lost 0
Inklingwon 18 · lost 0
Grok 4.3won 18 · lost 0
Qwen3.6 Pluswon 18 · lost 0
GPT-5.4 Miniwon 18 · lost 0
Gemini 3.5 Flash Litewon 18 · lost 0
Kimi K2.6won 18 · lost 0
Grok 4.5won 18 · lost 0
Claude Sonnet 4.6won 18 · lost 0
GLM-5.3 Flashwon 18 · lost 0
Qwen3.7 Maxwon 18 · lost 0
GLM-5.2won 18 · lost 0
DeepSeek V4 Flashwon 19 · lost 0
DeepSeek V4 Prowon 19 · lost 0
GLM-5.3won 19 · lost 0
Grok 4.6won 18 · lost 1
GPT-5.6 Lunawon 17 · lost 1
GPT-5.4won 17 · lost 1
Gemini 3.5 Flashwon 17 · lost 1
Gemini 3.6 Flashwon 17 · lost 1
Muse Spark 1.1won 17 · lost 1
GPT-5.2won 16 · lost 1
DeepSeek R1won 14 · lost 1
Gemini 2.5 Prowon 14 · lost 1
O4 Miniwon 13 · lost 1
O3won 13 · lost 1
GPT-5won 13 · lost 1
Kimi K2.5won 13 · lost 1
Mimo V2.6 Prowon 13 · lost 1
Gemini 3 Prowon 12 · lost 1
GLM-4.7won 11 · lost 1
GPT-5.1won 11 · lost 1
Grok 4.7won 16 · lost 2
Claude Sonnet 5won 16 · lost 2
GPT-5.6 Terrawon 16 · lost 2
Qwen3.8 Maxwon 16 · lost 2
Gemini 3.1 Prowon 16 · lost 2
Gemini 3.7 Flashwon 16 · lost 2
DeepSeek V4.1 Flashwon 16 · lost 3
Kimi K3won 16 · lost 3
Gemini 3.8 Flashwon 16 · lost 3
Claude Opus 4.8won 15 · lost 3
Muse Spark 1.2won 15 · lost 3
Claude Opus 4.6won 13 · lost 3
Claude Opus 4.7won 14 · lost 4
GPT-6 Solwon 13 · lost 5
GPT-5.6 Solwon 12 · lost 7
GPT-5.5won 11 · lost 7
GPT-6 Astrawon 10 · lost 9
Claude Opus 5won 8 · lost 11
Claude Opus 5.5won 5 · lost 13
Claude Fable 5won 4 · lost 14
Claude Fable 5.1won 4 · lost 15

§ 3 · Sources

Where the numbers come from

7 publications, 20 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1494
Artificial Analysis ↗1 measureread 2026-09-26
Artificial Analysis Intelligence Index 48
GDPval-AA ↗1 measureread 2026-09-26
GDPval-AA 1674
AA-Briefcase ↗1 measureread 2026-09-26
AA-Briefcase 1587
LiveBench ↗8 measuresread 2026-09-26
LiveBench 81.6LiveBench · Reasoning 89.7LiveBench · Coding 81.1LiveBench · Agentic Coding 64.1LiveBench · Mathematics 95.9LiveBench · Data Analysis 79.6LiveBench · Language 82.8LiveBench · Instruction Following 78
SimpleBench ↗1 measureread 2026-09-26
SimpleBench 81.8%
Vals.ai ↗7 measuresread 2026-09-26
Vals · Legal Research Bench 55.29%Vals · Harvey Legal Agent Benchmark 23.75%Vals · Finance Agent 59.96%Vals · CyberBench 72.74%Vals · Vibe Code Bench 85.86% (OpenHands)Vals · Code Migration 47.41%Vals · ProofBench 58%

Badge

PublicAI Index badge for Muse Spark 1.3[![PublicAI Index](https://publicai.io/model-index/badge?model=muse-spark-1-3)](https://publicai.io/model-index/m/muse-spark-1-3)