Muse Spark 1.3
Muse Spark 1.3 is strong in long context and reasoning; capable in agentic tasks, factuality, coding, math, and multimodal tasks; and behind the leaders in instruction following. Too few results yet to rate safety or multilingual tasks.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Safety or Multilingual.
Price
$1.25input$4.25outputper million tokens
From Artificial Analysis · 3 providers tracked · All prices
Evidence
54results on35benchmarks
- 18 independently verified
- 31 aggregator
- 5 vendor-reported
From 10 sources · latest Oct 21, 2026 · How verification works
API features
Tool callingStructured outputsReasoning
As listed by OpenRouter
Research
1 paper reference Muse Spark 1.3Muse Spark 1.3 benchmark results
54 results on 35 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
6.7% behind the leader1 of 3 ranked benchmarks measured
- 83.00Oct 16, 2026
9.3% behind the leader5 of 6 ranked benchmarks measured
- GPQA Diamond93.54Oct 16, 2026gpqa
- LiveBench · Reasoning89.65Oct 18, 2026livebench_reasoning@2026-06-25
- 81.80Sep 4, 2026
- 24.86Oct 16, 2026
- Humanity's Last Exam48.70Oct 16, 2026aa_hle
Show 3 more reasoning resultsHide 3 reasoning results
- 26.00Oct 16, 2026
- GPQA Diamond94.14Oct 16, 2026gpqa
- Humanity's Last Exam47.45Oct 16, 2026aa_hle
16.6% behind the leader4 of 7 ranked benchmarks measured
- τ-Bench V3 · Banking50.52Oct 16, 2026tauBanking
- 59.05Oct 16, 2026
- Terminal-Bench 4.033.33Oct 16, 2026
Show 4 more agentic resultsHide 4 agentic results
- 33.24Oct 16, 2026
- 56.36Oct 16, 2026
- Terminal-Bench 4.016.67Oct 16, 2026
- τ-Bench V3 · Banking47.22Oct 16, 2026tauBanking
17.3% behind the leader2 of 4 ranked benchmarks measured
- AA-Omniscience · Non-hallucination67.06Oct 16, 2026omniscienceNonHallucination
- AA-Omniscience · Accuracy43.58Oct 16, 2026omniscienceAccuracy
Show 2 more factuality resultsHide 2 factuality results
- AA-Omniscience · Accuracy41.52Oct 16, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination68.54Oct 16, 2026omniscienceNonHallucination
18.9% behind the leader5 of 10 ranked benchmarks measured
- Terminal-Bench 2.184.27Oct 16, 2026terminalbenchV21
- 1659.26Sep 21, 2026
- LiveBench · Coding81.06Oct 18, 2026livebench_coding@2026-06-25
- SciCode58.80Oct 16, 2026aa_scicode
- LiveBench · Agentic Coding64.09Oct 18, 2026livebench_agentic_coding@2026-06-25
Show 4 more coding resultsHide 4 coding results
- 1621.41Sep 21, 2026
- SciCode59.72Oct 16, 2026aa_scicode
- Terminal-Bench 2.185.39Oct 16, 2026terminalbenchV21
- 88.80Oct 20, 2026
21.1% behind the leader5 of 5 ranked benchmarks measured
- LiveBench · Mathematics95.95Oct 18, 2026livebench_math@2026-06-25
- 74.04Sep 21, 2026
- 46.34Sep 21, 2026
Show 2 more math resultsHide 2 math results
- 41.46Sep 21, 2026
- 74.39Sep 21, 2026
23.4% behind the leader1 of 6 ranked benchmarks measured
- MMMU-Pro82.02Oct 16, 2026aa_mmmu_pro
Show 2 more multimodal resultsHide 2 multimodal results
- 1301.82Oct 10, 2026
- 1308.41Sep 21, 2026
28.3% behind the leader1 of 3 ranked benchmarks measured
- LiveBench · Instruction Following78.00Oct 18, 2026livebench_instruction_following@2026-06-25
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- 66.89Oct 21, 2026
- livebench_language82.79Oct 18, 2026livebench_language@2026-06-25
- livebench_data_analysis79.57Oct 18, 2026livebench_data_analysis@2026-06-25
- 1494Oct 15, 2026
- AA Intelligence45.07Oct 16, 2026aa_intelligence_index
- 23.12Oct 16, 2026
Show 10 more resultsHide 10 results
- 51.83Sep 9, 2026
- 55.71Sep 9, 2026
- AA Intelligence48.09Oct 16, 2026aa_intelligence_index
- 25.00Oct 16, 2026
- Artificial Analysis Coding Index76.45Sep 9, 2026aa_coding_index
- Artificial Analysis Coding Index75.78Sep 9, 2026aa_coding_index
- DeepSearchQA (F1)89.40Oct 20, 2026DeepSearchQA
- 75.40Oct 20, 2026
- 64.90Oct 20, 2026
- 66.90Oct 20, 2026
Muse Spark 1.3: common questions
Who makes Muse Spark 1.3?
Muse Spark 1.3 is made by Meta.
When was Muse Spark 1.3 released?
Muse Spark 1.3 was released on Sep 2, 2026, according to Artificial Analysis.
What is Muse Spark 1.3 good at?
Muse Spark 1.3 is strong in long context and reasoning; capable in agentic tasks, factuality, coding, math, and multimodal tasks; and behind the leaders in instruction following. Too few results yet to rate safety or multilingual tasks.
How much does Muse Spark 1.3 cost?
Muse Spark 1.3 costs $1.25 per million input tokens and $4.25 per million output tokens, according to Artificial Analysis. We track its price at 3 providers. At a mix of three input tokens to one output token, it costs more than 72% of the 330 priced models we track.
How many benchmarks has Muse Spark 1.3 been tested on?
We track 54 results for Muse Spark 1.3 on 35 benchmarks from 10 sources, 18 of them independently verified. The latest was recorded on Oct 21, 2026.
Which API features does Muse Spark 1.3 support?
OpenRouter lists tool calling, structured outputs, and reasoning for Muse Spark 1.3.
About this record
Where Muse Spark 1.3's numbers come from, and every name it appears under.
- Tracked since
- Sep 2, 2026
- Newest source mention
- Oct 14, 2026
Where the results come from
Verification: 54 scores · 18 independently verified · 31 aggregator-attributed · 5 vendor-reported. How these tiers are assigned
From 10 sources on 8 sites. Artificial Analysis supplies 31 of them; the 18 independently verified results come from 6 sites. Bars are coloured by trust tier.
- artificialanalysis.ai31
- livebench.ai7
- api.llm-stats.com5
- datasets-server.huggingface.co4
- epoch.ai4
- labs.scale.com1
- lmarena.ai1
- simple-bench.com1
Also known as
How our sources name Muse Spark 1.3 at each reasoning setting.
| Setting | Short form | API id |
|---|---|---|
| xhigh | muse spark 1.3 (xhigh) | muse-spark-1-3-xhigh |
| max | muse spark 1.3 (max) | muse-spark-1.3-max |