Skip to content

Meta · Chat

Muse Spark 1.2

Meta 🇺🇸ChatMuse familyOne of its sources says the weights aren't publicly available.

Newer version: Muse Spark 1.3. We compared release dates and version names in our catalogue.

From Empiriolabs's listing:
Meta's updated frontier reasoning model with a 1,048,576-token context, image, video, audio, and PDF understanding, web search, and tool calling.
llminfo IQ
137.3
Maker's price
$0.125 blended / 1M
Max context
1M
Listings
28

Evidence and sources

llminfo IQ 137.3, ranked among chat models. From 199 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 116.1, 95% interval 110–122. From 18 benchmark families across 8 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedGood value (62.0). Cheaper per task than its benchmark results predict. From 37 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsChatty (66 ±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.Median 214,484 output tokens per task at max effort.How the Yap score worksReleased 5 Aug 2026Maker HQ: US (according to Meta's own website)
All rankings: #4 in Bluesky discussions this week #127 on OpenRouter today #149 on OpenRouter this week #187 on OpenRouter this month #235 on NanoGPT recently #235 on OpenRouter today #265 on OpenRouter this month #277 on OpenRouter this week #473 on NanoGPT recentlyAll rankings: #4 in Bluesky discussions this week #127 on OpenRouter today #149 on OpenRouter this week #187 on OpenRouter this month #235 on NanoGPT recently #235 on OpenRouter today #265 on OpenRouter this month #277 on OpenRouter this week #473 on NanoGPT recentlySources: LMArena data, EmpirioLabs data, Models.dev data, llminfo researchPrimary data from LMArena from EmpirioLabs: primary mode, description from Models.dev: weights availability from llminfo research: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 9 maker's prices; 1 market median price.

Input & output

as stated by the model's providers
Text
Images
Audio
Video
Files
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingPDF inputWeb searchAudio input

Listings

 Metameta/muse-spark-1.2-contributorSources: LiteLLM data, Models.dev data1M$0.10cache read $0.002$0.20——
 Kilometa/muse-spark-1.2-contributorSources: Kilo AI Gateway data, Models.dev data1M$0.10cache read $0.002$0.20——
 Llmgatewaymuse-spark-1.2-contributorSource: LLMGateway data1M$0.10cache read $0.002$0.20——
 NanoGPTmeta/muse-spark-1.2-contributorSources: Models.dev data, NanoGPT1M$0.10cache read $0.002$0.20——
 OpenRouterInference by Metameta/muse-spark-1.2-contributorSource: OpenRouter data docs1M$0.10cache read $0.002$0.201052,491ms
 Vercelmeta/muse-spark-1.2-contributorSources: Models.dev data, Vercel AI Gateway data1M$0.10cache read $0.002$0.20——
 Commandcodemeta/muse-spark-1.2-contributorSource: Models.dev data1M$0.10cache read $0.002$0.20——
 Llmtrmeta/muse-spark-1.2-contributorSource: Models.dev data1M$0.10cache read $0.002$0.20——
 Neosantaramuse-spark-1.2-contributorSource: Neosantara Inference data1M$0.10cache read $0.002$0.20——
 Opencode Goopencode_go/muse-spark-1.2-contributorSources: LiteLLM data, Models.dev data1M$0.10cache read $0.002$0.20——
Show all 25 listings

Showing 10 of 25: the maker's own listing first, then the cheapest per 1M input tokens from each provider.

Also sold by Abacus, Aihubmix, Aimlapi, Cloudflare AI Gateway, Eden AI, Empiriolabs, Fastrouter, Kenari, Llmgateway Providers, Nous Portal, Opper, Orcarouter, Zenmux.

3 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Arena ELO52 related benchmarks · 52 results
Source: LMArena data · Nov 2026
BenchmarkxHigh
Arena ELO
Chinese
Creative Writing
English
Show 48 more benchmarksHide 48 benchmarks
BenchmarkxHigh
Exclude Ties
Expert
French
German
Hard Prompts
Hard Prompts English
Industry Business and Management and Financial Operations
Industry Entertainment and Sports and Media
Industry Life and Physical and Social Science
Industry Mathematical
Industry Medicine and Healthcare
Industry Software and It Services
Industry Writing and Literature and Language
Japanese
Korean
Longer Query
Multi Turn
Non English
Polish
Russian
Spanish
Style-Controlled
Text Factuality
Text Factuality - Chinese
Text Factuality - Creative Writing
Text Factuality - English
Text Factuality - Exclude Ties
Text Factuality - Expert
Text Factuality - Hard Prompts
Text Factuality - Hard Prompts English
Text Factuality - Industry Business and Management and Financial Operations
Text Factuality - Industry Entertainment and Sports and Media
Text Factuality - Industry Life and Physical and Social Science
Text Factuality - Industry Mathematical
Text Factuality - Industry Medicine and Healthcare
Text Factuality - Industry Software and It Services
Text Factuality - Industry Writing and Literature and Language
Text Factuality - Longer Query
Text Factuality - Multi Turn
Text Factuality - Non English
Text Factuality - Russian
Text Factuality - Spanish
Vision
Vision - Diagram
Vision - English
Vision Style-Controlled
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Long context

GDM-MRCRv2: 8 needle8 related benchmarks · 40 results
Source: Context Arena data · Aug 2026
Benchmarkminimallowmediumhighxhigh
GDM-MRCRv2: 8 needle
8k
16k
32k
Show 4 more benchmarksHide 4 benchmarks
Benchmarkminimallowmediumhighxhigh
64k
128k
256k
512k
OpenAI MRCR v2 · 8-needlexhigh · 100 examples · 8-needle · no tools · mean sequence-matcher ratio · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Agentic

Agentic IF Indexxhigh · Meta internal composite · agentic instruction following · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Arena AgentSource: LMArena data
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026
GDPval-AA v22 results
Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
xhigh · Artificial Analysis Stirrup · 220 tasks · shell + web browsing · human baseline 1000
JobBenchxhigh · 65 tasks · official evaluation code · OpenCode · file-aware rubric grader · mean rubric score · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OSWorld 2.0 · Binaryxhigh · OSWorld 2.0 v06.24 · 108 tasks · common internal framework · binary completion · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OSWorld 2.0 · Partialxhigh · OSWorld 2.0 v06.24 · 108 tasks · common internal framework · partial reward · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SWE-Atlas · Codebase Q&Axhigh · public QnA split · 124 tasks · mini-swe-agent · rubric grading · mean pass@1 · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
xhigh · 89 tasks · native coding harness · isolated cloud sandbox · pass@1
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

General

Coding

Arena ELOSource: LMArena data · Nov 2026
Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
ValsSource: Vals AI

Speech

Audio MultiChallengeQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
FLEURS-60 · ASR WERQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MMARQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MMAUQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MMSUQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
VoiceBenchQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Multilingual

EKI EstonianSource: Keelemudelite mõõdupuu data docs · Aug 2026

Reasoning

AutomationBenchxhigh · public v3 · 600 tasks · benchmark automation tools · pass@1 · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSearchQAxhigh · 900 questions · common browser search/open/find harness · semantic answer matching · Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Math

Instruction following

Science

Factuality

Knowledge

OCR

Vision

Other

AISHELL-4 · cpWERQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
AISHELL-4 · DERQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
AliMeeting TestReported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
AVUTQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DailyOmniQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSWE v1.12 results
Reported by Meta · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
xhigh · official Datacurve leaderboard · task pass rate
HumMusQAQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
JoinAVBenchQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MagicData-RAMC · DERQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MLC-SLM (en) · cpWERQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MLC-SLM (en) · DERQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MuchoMusic-RULQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MusTBenchQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OmniCap-IF · CSRQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OmniCap-IF · ISRQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OmniClozeQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
QIVDQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SkillsBench · Vals6 related benchmarks · 6 results
With Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072
With Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072
With Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072
Without Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072
Without Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072
Without Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=meta/muse_spark_1_2 · provider=Meta · reasoning_effort=xhigh · temperature=1 · max_output_tokens=131072
SpotSoundBenchQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
StreamingBenchQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals14 related benchmarks · 16 results
Show 10 more resultsHide 10 results
LegalBench
Source: Vals AI
MedCode
Source: Clinical Benchmarks data docs · Sep 2026
Setting not stated
Source: Vals AI
Setting not stated
MedScribe
Source: Clinical Benchmarks data docs · Sep 2026
Setting not stated
Source: Vals AI
Setting not stated
WildSpeechQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
WorldSenseQwen3.8 Omni Flash 官方发布表 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated · BullshitBench by Peter Gostev, MIT (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · Data by LLM Stats · Keelemudelite mõõdupuu: Independent LLM leaderboard for Estonian. Eesti Keele Instituut. https://moodupuu.eki.ee, CC BY 4.0. Latest per-benchmark results selected with partial-run flags retained. (licence) · MineBench, MIT (licence)

Around the web

Also known as: aihubmix/muse-spark-1.2, empiriolabs/muse-spark-1-2, meta/muse-spark-1.2, meta/muse-spark-1.2-contributor, muse-spark-1-2, muse-spark-1-2-contributor, muse-spark-1.2-contributor, muse-spark-1.2-contributor-free, muse-spark-1.2-xhigh, opencode_go/muse-spark-1.2-contributor

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works