Code World Model (CWM) Benchmark Scores

Code World Model (CWM): Benchmark Scores as Published

Every result the catalog holds for Code World Model (CWM): 1 test in 1 area, from 1 page. Each was printed by the company named beside it. Each number is as printed; nothing is ranked or rescored.

Developer MetaReleased 2025-09-24Weights OpenData version 2026103001, as of 2026-10-30

Agentic Coding

Test	Value	Printed by	Setup as printed
SWE-bench Verified	53.9%	Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15	-

More from Meta

Muse Spark Open
Muse Wonder
Muse Wonder Full
Muse Glimmer-30B
Muse Spark 1.2
Llama 4 Maverick
Llama 4 Maverick (pre-trained)
Llama 4 Scout
Llama 4 Scout (pre-trained)
Llama 3.3 70B
Llama 3.1 405B
Llama 3.1 405B (pre-trained)
Llama 3.1 70B (pre-trained)
Muse Spark 1.3

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.