Claude Sonnet 5 vs Muse Spark 1.3

Anthropic against Meta. Prices, scores and speed below are live.

  • Muse Spark 1.3 costs 50% less on blended price ($3.00 against $6.00 per 1M tokens, 3 input to 1 output).
  • Claude Sonnet 5 scores higher on the Artificial Analysis Intelligence Index: 38.4 against 36.6.
  • On the 35 benchmarks both have scores on, Claude Sonnet 5 leads on 3 and Muse Spark 1.3 on 32.
  • Muse Spark 1.3 reads more context: 1M tokens against 1M.
  • Muse Spark 1.3 is newer: listed Sep 2, 2026, against Jun 30, 2026.

Side by side

Claude Sonnet 5Muse Spark 1.3
LabAnthropicMeta
ListedJun 30, 2026Sep 2, 2026
Input per 1M tokens$3.00$1.88
Output per 1M tokens$15.00$6.38
Cached input per 1M$0.30$0.23
Blended (3:1) per 1M$6.00$3.00
Context window1M1M
Max output128K944K
Knowledge cutoff--
Intelligence index38.436.6
Coding index71.575.8
ReasoningYesYes
Reads imagesYesYes
Tool useYesYes
Structured outputYesYes
Measured output speed70 tokens/s-

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

BenchmarkClaude Sonnet 5Muse Spark 1.3
AIME80%Epoch AI99.2%Epoch AI
APEX-Agents54.5%Epoch AI58.6%Epoch AI
Artificial Analysis Agentic Index43.6Artificial Analysis and OpenRouter55.5Artificial Analysis and OpenRouter
Artificial Analysis Coding Index71.5Artificial Analysis and OpenRouter75.8Artificial Analysis and OpenRouter
Artificial Analysis Intelligence Index38.4Artificial Analysis and OpenRouter36.6Artificial Analysis and OpenRouter
Chess Puzzles45%Epoch AI37%Epoch AI
Design Arena: 3D1258 EloDesign Arena1394 EloDesign Arena
Design Arena: Android Native (agentic)1221 EloDesign Arena1288 EloDesign Arena
Design Arena: ASCII Art1204 EloDesign Arena1288 EloDesign Arena
Design Arena: Code1274 EloDesign Arena1350 EloDesign Arena
Design Arena: Data Viz1252 EloDesign Arena1342 EloDesign Arena
Design Arena: Full-stack (agentic)1249 EloDesign Arena1312 EloDesign Arena
Design Arena: Game Dev1275 EloDesign Arena1344 EloDesign Arena
Design Arena: Godot Game Dev (agentic)1225 EloDesign Arena1241 EloDesign Arena
Design Arena: Mobile Apps (agentic)1208 EloDesign Arena1221 EloDesign Arena
Design Arena: PPTX Slides (agentic)1169 EloDesign Arena1201 EloDesign Arena
Design Arena: SVG1190 EloDesign Arena1322 EloDesign Arena
Design Arena: UI Component1274 EloDesign Arena1347 EloDesign Arena
Design Arena: Web Apps (agentic)1227 EloDesign Arena1282 EloDesign Arena
Design Arena: Website1273 EloDesign Arena1356 EloDesign Arena
DTBench92.5%Epoch AI96.5%Epoch AI
FrontierMath Tier 4 (v2)29.3%Epoch AI46.3%Epoch AI
FrontierMath Tiers 1-3 (v2)65.6%Epoch AI74%Epoch AI
LMArena (Coding)1484 EloLMArena1509 EloLMArena
LMArena (Overall)1443 EloLMArena1491 EloLMArena
LMArena: Creative Writing1416 EloLMArena1464 EloLMArena
LMArena: Hard Prompts1462 EloLMArena1503 EloLMArena
LMArena: Longer Query1463 EloLMArena1488 EloLMArena
LMArena: Math1482 EloLMArena1492 EloLMArena
LMArena: Multi-Turn1453 EloLMArena1488 EloLMArena
LMCA50Epoch AI53.9Epoch AI
Mystery Game Puzzles16%Epoch AI25%Epoch AI
ProofBench77%Epoch AI58%Epoch AI
SciCode54.3%Epoch AI59.7%Epoch AI
WebDev Arena1537 EloEpoch AI1652 EloEpoch AI

In our tests and guides

Where Claude Sonnet 5 or Muse Spark 1.3 ran on a real brief, or shows in a live table.

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See Anthropic models, Meta models, every LLM's price and our sources.