Search “Meta”

English translation is being prepared.

Meta releases Muse Wonder and Muse Wonder Full, surpassing Claude on long-horizon coding tests…

Meta Superintelligence Labs launched its new-generation flagship models Muse Wonder and Muse Wonder Full on October 13. The Full version scored 56 on the Artificial Analysis Composite Intelligence Index, 2 points behind Claude Opus 5.5's 58. On the DeepSWE 1.1 long-horizon software engineering test, Wonder Full scored 79.0, surpassing Opus 5.5's 74.2 and GPT-6 Astra's 74.1, but it still trails Opus on Terminal-Bench 4.0 and ProgramBench. Artificial Analysis estimates its average task cost at $4.52, up 183% from the previous generation Muse Spark 1.3; a developer comparison on the same prompt shows its 3D game generation is noticeably worse than Opus, and Meta's model card has acknowledged 3D tasks as a known limitation.

Meta releases the Muse Wonder series of models, focused on personal agent memory and long-horizon tasks

Meta released two models, Muse Wonder and Muse Wonder Full, internally codenamed Watermelon, targeting personal and professional agent tasks, with a focus on improving cross-conversation memory and long-duration task execution. Artificial Analysis evaluations show that Wonder Full placed in the top three on six of ten tests, and at the highest reasoning intensity, Muse Code paired with Wonder Full matches Claude Code paired with Sonnet 5.5 in coding performance. API pricing is $4 per million input tokens and $20 per million output tokens; Wonder Full is available only to Muse Code subscribers, and Meta AI head Alexandr Wang acknowledged that its coding capability still lags behind Opus 5.5.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

Meta open-sources 30B model Muse Glimmer, runs locally on 18GB of memory

Meta open-sources Muse Glimmer, a 30B-parameter dense model that runs on local devices with 18GB of memory, uses the Apache 2.0 license and supports vision. Mark Zuckerberg calls it the strongest agent model at its size and says the weights of the latest foundation model Muse Spark 1.2 will also be open-sourced later. Unsloth already provides a GGUF version of the model along with running and training guides.

English translation is being prepared.

Meta's Muse Glimmer 30B Launches on Fireworks, Targeting Persistent Agents

Fireworks has launched Meta's Muse Glimmer 30B dense model under the Apache 2.0 license, supporting 128K+ token context and native image understanding, targeting persistent agents that require multi-turn tool calling. The model scores 75.5 on MCP Atlas, 74.6 on DeepSearch QA, and 51.2 on SWE-Bench Pro, all higher than the corresponding results of Gemma 4 31B and Qwen 3.6 27B; these benchmark figures are reported by Meta. Fireworks also provides recommended settings of temperature 1.0, top_p 0.95, and top_k 64, and supports adjusting reasoning intensity in the system prompt.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

Microsoft MAI-Image-2.7-Preview Rises to Second on the Image Editing Leaderboard

Mustafa Suleyman announced that MAI-Image has become the world's second-ranked image editing model, saying it once again surpasses Google, Meta, xAI and GPT-Image-2.5-Flare, and that it is available to try in MAI Playground. The Artificial Analysis evaluation he cited shows that Microsoft MAI-Image-2.7-Preview ranks second on AA-Image-Editing v2.1, behind only OpenAI's GPT Image 2.5 Sunburst (max), scoring higher than MAI-Image-2.6 on all 7 editing actions, including gains of +57 Elo in composition and framing, +49 in text or symbol editing, and +48 in scene and style editing, and ranking first on 4 of the actions. The model currently offers image editing only in MAI Playground, with text-to-image functionality said to be coming soon.