Why does every AI lab feel the need to build their own coding agent…? Don’t we have more than enough already?
mchusma
This is a nice release and a solid improvement over Spark 1.1. It compares favorably with Grok 4.5. Not SOTA, but solid releases. I think they need to really get this more competitive with Deepseek V4 Flash / Luna pricing to move the needle.
Interesting that they have separate API pricing for "we can train on your data" (whereas iirc most of the big players either make that distinction only between subscriptions and API usage, or train on everything). Wonder how it compares to Deepseek V4 Flash given that they're similar on pricing and data policy.
Very interesting they have a way cheaper "contributor" version "used to improve our products", how much of that is price discrimination vs the data being that valuable?
Roughly DeepSeek V4 Flash pricing, though you can get V4 from providers that don't train on your data
ipsum2
I wonder why they didn't compare with GPT-5.6-sol, only Terra?
show comments
giancarlostoro
Will someone at Meta for the love of God make it so none of this stuff goes through Facebook.com? You want customers but most corporate firewalls block social media. Also, a lot of devs do not want their work stuff tied up to their facebook account. For the love of all things show the IG / FB logins as optional and do email as primary.
I am not a fan of Meta but I do cheer for any competitors against OpenAI and Anthropic, the duopoly is getting tiresome.
show comments
liviux
Does this muse code have any muse spark 1.2 usage included? Can't understand from the docs.
minimaxir
Muse Spark 1.1 was released July 16th, less than a month ago. A new version release this soon (particularly after Kimi K3's release drastically overshadowed it) is a bit sus and it appears that Meta is trying a first launch do-over.
fcoury
Interesting, it seems like their muse code is built upon Codex CLI?
Bolwin
> Muse Spark 1.2 is available today in Muse Code and in Meta Model API with expanded global access
Wasn't the previous one us only? This is probably the biggest part of the post
Anyone know if muse code is open source?
kcb
Open the weights.
paulkrush
Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows. In Muse Spark 1.2, we significantly scaled up training compute on coding tasks while expanding training environment diversity. The model also maintains its strength in other key areas like general agents.
rvz
First of all, you have login to use it. Why?
After everything that you have seen with Meta, would you really trust them with a coding agent? You don't even know if your prompts are being analyzed by them on the side or if your code base is being uploaded to them. This goes for the rest of them that have closed harnesses and closed models gated by a login.
Think twice before falling for this announcement and ask yourself what they are not telling you.
esafak
If anyone from Meta is reading, please can you publish the cost and latency for each of your benchmarks, like OpenAI does? This needs to become standard practice.
qphe95
Theres no actual evidence they didn't just distill Kimi K3
Why does every AI lab feel the need to build their own coding agent…? Don’t we have more than enough already?
This is a nice release and a solid improvement over Spark 1.1. It compares favorably with Grok 4.5. Not SOTA, but solid releases. I think they need to really get this more competitive with Deepseek V4 Flash / Luna pricing to move the needle.
Pricing: https://dev.meta.ai/docs/pricing-rate-limits
Interesting that they have separate API pricing for "we can train on your data" (whereas iirc most of the big players either make that distinction only between subscriptions and API usage, or train on everything). Wonder how it compares to Deepseek V4 Flash given that they're similar on pricing and data policy.
https://pbs.twimg.com/media/HO-59jQaoAA_JZ1?format=jpg
Very interesting they have a way cheaper "contributor" version "used to improve our products", how much of that is price discrimination vs the data being that valuable?
Roughly DeepSeek V4 Flash pricing, though you can get V4 from providers that don't train on your data
I wonder why they didn't compare with GPT-5.6-sol, only Terra?
Will someone at Meta for the love of God make it so none of this stuff goes through Facebook.com? You want customers but most corporate firewalls block social media. Also, a lot of devs do not want their work stuff tied up to their facebook account. For the love of all things show the IG / FB logins as optional and do email as primary.
I am not a fan of Meta but I do cheer for any competitors against OpenAI and Anthropic, the duopoly is getting tiresome.
Does this muse code have any muse spark 1.2 usage included? Can't understand from the docs.
Muse Spark 1.1 was released July 16th, less than a month ago. A new version release this soon (particularly after Kimi K3's release drastically overshadowed it) is a bit sus and it appears that Meta is trying a first launch do-over.
Interesting, it seems like their muse code is built upon Codex CLI?
> Muse Spark 1.2 is available today in Muse Code and in Meta Model API with expanded global access
Wasn't the previous one us only? This is probably the biggest part of the post
Anyone know if muse code is open source?
Open the weights.
Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows. In Muse Spark 1.2, we significantly scaled up training compute on coding tasks while expanding training environment diversity. The model also maintains its strength in other key areas like general agents.
First of all, you have login to use it. Why?
After everything that you have seen with Meta, would you really trust them with a coding agent? You don't even know if your prompts are being analyzed by them on the side or if your code base is being uploaded to them. This goes for the rest of them that have closed harnesses and closed models gated by a login.
Think twice before falling for this announcement and ask yourself what they are not telling you.
If anyone from Meta is reading, please can you publish the cost and latency for each of your benchmarks, like OpenAI does? This needs to become standard practice.
Theres no actual evidence they didn't just distill Kimi K3