BlogLLM Benchmark Scores as Published, Side by Side

Muse Spark 1.2: Benchmark Scores as Published

Every result the catalog holds for Muse Spark 1.2: 6 tests in 4 areas, from 2 pages. Each was printed by the company named beside it. Each number is as printed; nothing is ranked or rescored.

Developer MetaReleased 2026-08-05Weights ClosedData version 2026103001, as of 2026-10-30

Knowledge Work

TestValuePrinted bySetup as printed
GDPval-AA v21628Google, Gemini 3.7 Flash model cardchecked 2026-10-15-

Agentic Coding

TestValuePrinted bySetup as printed
Terminal-Bench 2.182.9%Google, Gemini 3.7 Flash model cardchecked 2026-10-15-
DeepSWE v1.154.9%Google, Gemini 3.7 Flash model cardchecked 2026-10-15-
Code Arena1535Google, Gemini 3.7 Flash model cardchecked 2026-10-15-

Multimodal Understanding

TestValuePrinted bySetup as printed
GDP.pdf16%Google, Gemini 3.7 Flash model cardchecked 2026-10-15-

Preference and Composite Rankings

TestValuePrinted bySetup as printed
Artificial Analysis Intelligence Index57Google, Gemini 3.7 Flash model cardchecked 2026-10-15-

Price

API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked.

PriceValuePrinted bySetup as printed
Input$1.25Meta, Meta Model API pricingchecked 2026-10-19-
Cached input$0.15Meta, Meta Model API pricingchecked 2026-10-19-
Output$4.25Meta, Meta Model API pricingchecked 2026-10-19-
Input, contributor$0.1Meta, Meta Model API pricingchecked 2026-10-19-
Cached input, contributor$0.002Meta, Meta Model API pricingchecked 2026-10-19-
Output, contributor$0.2Meta, Meta Model API pricingchecked 2026-10-19-

More from Meta

Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.