---
format: "aidr-story-markdown/v1"
id: "7540007e2f30a94c6b8610d635d916528557409b57bde5359c474360b6ea703c"
canonical_url: "https://aidr.today/7540007e?lang=en"
title: "Meta Releases Muse Spark 1.3 Tied With GPT-5.6 Sol at 61 and Cheapest Above 59 on Artificial Analysis Index"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-03T11:48:04.000Z"
category: "Models"
topics: ["meta","llm","benchmark","coding"]
source_urls: ["https://huggingnews.com/ai/meta-releases-muse-spark-13-tied-with-gpt-56-sol-at-61-and-cheapest-abov-c1a18141","https://x.com/jilin_14/status/2095370086345564581","https://x.com/WesRoth/status/2095368689763619309","https://x.com/WesRoth/status/2095383789392236614","https://x.com/kimmonismus/status/2095466152696668202","https://x.com/eliebakouch/status/2095393673492009166","https://x.com/shishirpatil_/status/2095307695050178810","https://x.com/ArthurMacwaters/status/2095315716442595748"]
summary: "Developers can now access Muse Spark 1.3 in Muse Code and the Meta Model API, and OpenCode separately listed the model as free. The generally available xhigh version scored 61 on the Artificial Analysis Intelligence Index, matching GPT-5.6 Sol, and cost $0.55 per task versus $0.95 for Sol. That made it the lowest-cost model above 59 on the index. The max version, in limited preview for partners, scored 62. Other benchmark figures showed DeepSWE at 75.4, up from 55.0, and Terminal-bench 2.1 at 88.8, tied with GPT-5.6 Sol. An MRCR long-context score of 98.1, up from 55.5, drew pushback from one researcher who said the benchmark can be overfit and that Muse Spark 1.3 did not appear better on other long-context tests."
---

# Meta Releases Muse Spark 1\.3 Tied With GPT\-5\.6 Sol at 61 and Cheapest Above 59 on Artificial Analysis Index

> [Open the canonical story](<https://aidr.today/7540007e?lang=en>)

**Published:** 2026-09-03T11:48:04.000Z
**Category:** Models
**Topics:** meta, llm, benchmark, coding

## Summary

Developers can now access Muse Spark 1\.3 in Muse Code and the Meta Model API, and OpenCode separately listed the model as free\. The generally available xhigh version scored 61 on the Artificial Analysis Intelligence Index, matching GPT\-5\.6 Sol, and cost $0\.55 per task versus $0\.95 for Sol\. That made it the lowest\-cost model above 59 on the index\. The max version, in limited preview for partners, scored 62\. Other benchmark figures showed DeepSWE at 75\.4, up from 55\.0, and Terminal\-bench 2\.1 at 88\.8, tied with GPT\-5\.6 Sol\. An MRCR long\-context score of 98\.1, up from 55\.5, drew pushback from one researcher who said the benchmark can be overfit and that Muse Spark 1\.3 did not appear better on other long\-context tests\.

## Sources

- [Story source](<https://huggingnews.com/ai/meta-releases-muse-spark-13-tied-with-gpt-56-sol-at-61-and-cheapest-abov-c1a18141>)
- [Story source](<https://x.com/jilin_14/status/2095370086345564581>)
- [Supporting source](<https://x.com/WesRoth/status/2095368689763619309>)
- [Supporting source](<https://x.com/WesRoth/status/2095383789392236614>)
- [Supporting source](<https://x.com/kimmonismus/status/2095466152696668202>)
- [Supporting source](<https://x.com/eliebakouch/status/2095393673492009166>)
- [Story source](<https://x.com/shishirpatil_/status/2095307695050178810>)
- [Supporting source](<https://x.com/ArthurMacwaters/status/2095315716442595748>)