Updated 2026-09-05 10:32 UTC
Meta news
Everything Meta has published that we can see: blog posts, release notes, changelog entries, incidents and deprecations. Their words, linked back to them, newest first.
What Meta has shipped lately
Mon 27 Jul 2026
1Tue 21 Jul 2026
1How Meta’s AI Models Are Powering the First Wave of Genesis Mission Projects
Thu 09 Jul 2026
1Introducing Muse Spark 1.1
Tue 07 Jul 2026
1Introducing Muse Image and Muse Video
Muse Image follows instructions faithfully, edits with precision, composes from multiple references, and draws on Instagram for social context. Muse Video delivers exceptional visual fidelity with native audio support.
Mon 29 Jun 2026
1From Brain Waves to Words: Brain2Qwerty Offers a New Path to Communication Without Surgery
Wed 08 Apr 2026
2Introducing Muse Spark: Scaling Towards Personal Superintelligence
Scaling How We Build and Test Our Most Advanced AI
As we build more capable, personalized AI, reliability, security, and user protections are more important than ever.
Mon 06 Apr 2026
1How Alta Daily Uses Meta’s Segment Anything to Reimagine the Digital Closet
Thu 26 Mar 2026
1Introducing TRIBE v2: A Predictive Foundation Model Trained to Understand How the Human Brain Processes Complex Stimuli
Tue 10 Mar 2026
1Mapping the World's Forests with Greater Precision: Introducing Canopy Height Maps v2
In partnership with the World Resources Institute, today we’re announcing Canopy Height Maps v2 (CHMv2), an open source model, along with world-scale maps generated with the model.
Wed 12 Nov 2025
1Model repo updated: meta-llama/Prompt-Guard-86M
text classification · 4,496,208 downloads · 397 likes · tags: deberta-v2, text-classification, facebook, meta, llama, llama-3, en, text-embeddings-inference
Wed 18 Jun 2025
2Model repo updated: meta-llama/Meta-Llama-3-8B-Instruct
text generation · 1,492,731 downloads · 4,910 likes · tags: llama, text-generation, facebook, meta, llama-3, conversational, en, eval-results
Model repo updated: meta-llama/Meta-Llama-3-70B-Instruct
text generation · 59,486 downloads · 1,525 likes · tags: llama, text-generation, facebook, meta, llama-3, conversational, en, eval-results
Thu 22 May 2025
3Model repo updated: meta-llama/Llama-4-Maverick-17B-128E-Instruct
image text to text · 11,338 downloads · 505 likes · tags: llama4, image-text-to-text, facebook, meta, llama, conversational, ar, de
Model repo updated: meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8
image text to text · 70,254 downloads · 176 likes · tags: llama4, image-text-to-text, facebook, meta, llama, conversational, ar, de
Model repo updated: meta-llama/Llama-4-Scout-17B-16E-Instruct
image text to text · 174,923 downloads · 1,339 likes · tags: llama4, image-text-to-text, facebook, meta, llama, conversational, ar, de
Sat 10 May 2025
1Model repo updated: meta-llama/Llama-4-Maverick-17B-128E-Instruct-Original
3 downloads · 40 likes · tags: facebook, meta, llama, llama-4, ar, de, en, es
Tue 29 Apr 2025
3Model repo updated: meta-llama/Llama-Guard-4-12B
image text to text · 61,096 downloads · 121 likes · tags: llama4, image-text-to-text, facebook, meta, llama, safety, conversational, en
New model repo: meta-llama/Llama-Prompt-Guard-2-86M
text classification · 104,820 downloads · 192 likes · tags: deberta-v2, text-classification, facebook, meta, llama, llama4, safety, en
New model repo: meta-llama/Llama-Prompt-Guard-2-22M
text classification · 11,224 downloads · 60 likes · tags: deberta-v2, text-classification, facebook, meta, llama, llama4, safety, en
Showing the newest 20 of 20 items we hold for Meta. RSS for Meta only · every lab together
Meta's current line-up
| Model | Released | Context | In / 1M | Out / 1M |
|---|---|---|---|---|
| Muse Spark 1.3 | 2026-09-02 | 1M | $1.25 | $4.25 |
| Muse Glimmer | 2026-08-10 | 131.1K | — | — |
| Muse Spark 1.1 | 2026-07-09 | 1M | $1.25 | $4.25 |
| Muse Spark 1.2 | — | 1M | $1.25 | $4.25 |
| Llama 4 Scout | — | — | — | — |
Released and preview models from Meta's own docs. A dash means they haven't published that number.
Meta prices run from $1.25 to $1.25 per million input tokens across 3 models. The Meta price page works out what your own usage would cost on each one.
Where this comes from
We read Meta's own feeds every fifteen minutes and store what's new. Nothing here is rewritten, and nothing comes from a rumour account or a reposted screenshot.
Sources: meta-ai-blog, hf-meta-llama.