BenchLeader

MiMo-V2.6-Pro vs Muse Spark 1.3

Full profiles: MiMo-V2.6-Pro (Xiaomi) and Muse Spark 1.3 (Meta): every benchmark, price and rank.

Updated · Data from 11 sources

As of 16 Nov 2026, MiMo-V2.6-Pro and Muse Spark 1.3 are neck and neck on quality (57.5 vs 61.6 on the BenchLeader Index, inside the ±9.1 margin), and MiMo-V2.6-Pro is 3.7× cheaper at $0.544 vs $2.00 per 1M tokens blended. Choose MiMo-V2.6-Pro for maths (66 vs 56); choose Muse Spark 1.3 for fast first answers (2.6 s vs 42.3 s) and knowledge (81 vs 42), reasoning (71 vs 50), and agents & tools (61 vs 57), where it also streams ~1.5× faster.

Key facts (as of 16 Nov 2026)

  • MiMo-V2.6-Pro scores 57.5 on the BenchLeader Index and Muse Spark 1.3 scores 61.6 (BenchLeader, 16 Nov 2026).
  • MiMo-V2.6-Pro costs $0.544 per 1M tokens blended (list price, 16 Nov 2026).
  • Muse Spark 1.3 costs $2.00 per 1M tokens blended (list price, 16 Nov 2026).
  • Muse Spark 1.3 streams ~1.5× faster than MiMo-V2.6-Pro (Artificial Analysis, 7-day median, 16 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
Agents & tool useMuse Spark 1.361 vs 57 category score; APEX-Agents 58.6% vs 54.4%Medium
MathsMiMo-V2.6-Pro66 vs 56 category score; ProofBench 70.0% vs 55.0%High
ReasoningMuse Spark 1.371 vs 50 category scoreHigh
KnowledgeMuse Spark 1.381 vs 42 category scoreHigh
Chat quality (human preference)–Only MiMo-V2.6-Pro has an Arena rating (67 category score)Not comparable
Lowest costMiMo-V2.6-Pro$0.548 vs $2.00 per 1M blended; agentic coding session $0.030 vs $0.092–
Fastest first answerMuse Spark 1.32.6 s (Muse Spark 1.3) vs 42.3 s (MiMo-V2.6-Pro) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • MiMo-V2.6-Pro and Muse Spark 1.3 are neck and neck on quality (57.5 vs 61.6, within the ±9.1 margin).
  • MiMo-V2.6-Pro is stronger in maths.
  • Muse Spark 1.3 is stronger in agents & tools, knowledge, reasoning.
  • MiMo-V2.6-Pro is 3.7× cheaper ($0.544 vs $2.00 per 1M blended).
  • Muse Spark 1.3 streams 3.0× faster (163 vs 54 tokens per second).

At a glance

SpecMiMo-V2.6-ProMuse Spark 1.3
DeveloperXiaomiMeta
Released21 Sep 20262 Sep 2026
Context window (tokens)1,048,5761,048,576
Compared atMiMo-V2.6-ProMuse Spark 1.3
Open weightsYesNo
Price in / out / cached input per 1M$0.440 / $0.870 / $0.0036$1.25 / $4.25 / $0.150

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryMiMo-V2.6-ProMuse Spark 1.3Leader
Reasoning49.971.0Muse Spark 1.3
Coding58.8–not comparable
Agents & tools57.161.3Muse Spark 1.3
Maths65.556.3MiMo-V2.6-Pro
Knowledge42.381.2Muse Spark 1.3
Human preference67.0–not comparable
Multimodal60.2–not comparable
Long context67.9–not comparable

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningAgents & toolsMathsKnowledgeMiMo-V2.6-Pro, Reasoning: 49.9MiMo-V2.6-Pro, Agents & tools: 57.1MiMo-V2.6-Pro, Maths: 65.5MiMo-V2.6-Pro, Knowledge: 42.3Muse Spark 1.3, Reasoning: 71.0Muse Spark 1.3, Agents & tools: 61.3Muse Spark 1.3, Maths: 56.3Muse Spark 1.3, Knowledge: 81.2

Not drawn, measured for fewer than two of these models: Coding, Instruction following, Long context, Multimodal, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 2 benchmarks both have, MiMo-V2.6-Pro leads 1, Muse Spark 1.3 leads 1. Every overall index that rates both puts Muse Spark 1.3 ahead.

Overall indexMiMo-V2.6-ProMuse Spark 1.3Leader
BenchLeader Index57.561.6neck and neck
AA Intelligence Index v5.0.129.936.6†Muse Spark 1.3
Arena Text14801497†Muse Spark 1.3

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureMiMo-V2.6-ProMuse Spark 1.3
Blended price per 1M tokens$0.548$2.00
Input / output per 1M$0.440 / $0.870$1.25 / $4.25
Cached input per 1M$0.0036$0.150
Price per task: chat reply$0.0004$0.0018
Price per task: summarize a 30-page report$0.0058$0.018
Price per task: code edit$0.0039$0.014
Price per task: agentic coding session$0.030$0.092
Price per task: structured extraction$0.0011$0.0034
Agentic coding session, input mostly cached$0.010$0.043
Cost per benchmark task (AA Index)$0.257–
Output tokens per task (AA Index)121k–
Of which reasoning86k–
Time per task (AA Index)38 min–
Output speed54 tok/s163 tok/s
Time to first answer42.3 s2.6 s
Time to a complete answer, last measured 13 Nov 202695.7 s–

Each model is compared at its best-ranked reasoning-effort setting. MiMo-V2.6-Pro scores 57.5 and takes 42 s to its first answer token. Muse Spark 1.3 scores 61.6 and takes 3 s to its first answer token; Muse Spark 1.3 (max) scores 60.6 in 31 s, Muse Spark 1.3 (xhigh) scores 58.1 in 11 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

137 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherMiMo-V2.6-ProMuse Spark 1.3
BenchLeader IndexBenchLeader57.561.6
Agents & tools scoreBenchLeader57.161.3
Coding scoreBenchLeader58.8–
Human preference scoreBenchLeader67.0–
Knowledge scoreBenchLeader42.381.2
Long context scoreBenchLeader67.9–
Maths scoreBenchLeader65.556.3
Multimodal scoreBenchLeader60.2–
Reasoning scoreBenchLeader49.971.0
Blended price $/MBenchLeader$0.548$2.00
Output speedBenchLeader54 tok/s163 tok/s
Time to first answerBenchLeader42.3 s2.6 s
Context windowBenchLeader1,048,576 tokens1,048,576 tokens
Reasoning
SimpleBenchSimpleBench–81.8%
Arena Hard PromptsArena (LMArena)15041517†
LiveBench 2.0 ReasoningLiveBench79.6%†89.7%†
LiveBench 1.0 ReasoningLiveBench–89.7%†
GPQA Diamond (AA)Artificial Analysis–93.5%†
Humanity's Last Exam (AA)Artificial Analysis49.4%48.7%†
HLE-Diamond (AA)Artificial Analysis15.2%33.7%†
CritPtArtificial Analysis26.6%24.9%†
CritPt v2 hard subset (AA)Artificial Analysis34.0%48.0%†
MysteryMechanismVals AI15.3%36.0%†
HLE-Diamond (Scale)Scale AI SEAL–25.4%
Arena ExpertArena (LMArena)15261521†
Chess PuzzlesEpoch AI Benchmarking Hub–37.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub–25.0%†
LMCAEpoch AI Benchmarking Hub–53.9%
DTBenchEpoch AI Benchmarking Hub–96.5%
ForecastBenchForecastBench–65.5%†
BTF-3Epoch AI Benchmarking Hub–14.1%†
Coding
SciCodeSciCode60.9%58.8%†
Arena CodingArena (LMArena)15381534†
Arena WebDevArena (LMArena)16251655†
LiveBench 2.0 CodingLiveBench64.5%†84.4%†
LiveBench 1.0 CodingLiveBench–81.1%†
IOIVals AI39.3%56.6%†
Code MigrationVals AI43.0%47.4%†
Show all 137 benchmarks
MetricPublisherMiMo-V2.6-ProMuse Spark 1.3
Coding
Code Migration v1.1Vals AI48.8%–
ProgramBenchVals AI0.5%2.5%†
SWE-bench MultimodalVals AI41.3%43.1%†
Vibe Code Bench 1-100Vals AI–20.5%†
Vibe Code Bench v1.1Vals AI85.2%85.9%†
Vibe Code Bench 1-100 v1.1Vals AI26.4%–
MirrorCodeEpoch AI Benchmarking Hub–54.4%†
CursorBenchCursor–41.6%†
ALE-BenchEpoch AI Benchmarking Hub1158.11688.3†
FrontierSWEEpoch AI Benchmarking Hub–25.8%†
WeirdML v3WeirdML–6.7%†
Terminal-Bench 4.0 (AA)Artificial Analysis34.9%33.3%†
Terminal-Bench 2.1 (AA)Artificial Analysis–84.3%†
Startrise FrontendStartrise–74.7%
GameBench 2 OverallDuelLab GameBench 2–57.7
GameBench 2 by reasoning settingDuelLab GameBench 249.8†68.5†
Design Arena WebsiteDesign Arena13111353†
FrontierCode 1.1 (AA)Artificial Analysis45.4%40.6%†
FrontierCode 2.0 (AA)Artificial Analysis24.0%23.1%†
Agents & tools
Terminal-BenchTerminal-Bench27.0%†14.8%†
APEX-AgentsMercor54.4%58.6%
Arena AgentArena (LMArena)2.42.2†
LiveBench Long AgenticLiveBench36.4%†47.9%†
LiveBench BrowserLiveBench77.3%†30.9%†
LiveBench Agentic CodingLiveBench–64.1%†
Terminal-Bench 2.1 (Vals)Vals AI67.8%79.0%†
HiL-BenchScale AI SEAL–66.9%
GDPval-AA v2.1, normalised scoreArtificial Analysis59.1%–
GDPval-AA v2.1Artificial Analysis16921681†
τ³-Banking (AA)Artificial Analysis–50.5%†
ITBench SRE (AA)Artificial Analysis–33.2%†
Analyst Agent (AA)Artificial Analysis37.5%37.5%†
CUA-benchVals AI–5.8%†
CyberBenchVals AI72.9%72.7%†
Finance Agent v2Vals AI57.3%60.0%†
Finance Agent v2.1Vals AI64.3%–
Harvey's Legal Agent Benchmark v1.1Vals AI13.8%–
Harvey's Legal Agent Benchmark v1Vals AI10.8%23.8%†
Legal Research BenchVals AI47.1%55.3%†
SREBenchVals AI3.0%3.8%†
Tax Agent BenchVals AI32.4%42.4%†
Terminal-Bench 4.0 (Vals)Vals AI31.3%24.8%†
Terminal-Bench ScienceVals AI2.9%10.0%†
Arena Agent Task CompletionArena (LMArena)53.7†
AutomationBenchArtificial Analysis58.6%57.9%†
GDP.pdf (AA)Artificial Analysis19.2%26.6%†
Harvey LAB-AA v1.1Harvey–8.9%†
AA-Briefcase v1.1Artificial Analysis15121581†
Design Arena AgenticDesign Arena12711283†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis5.7%10.9%†
CWE-Bench (AA)Artificial Analysis63.3%60.8%†
DeepSec-Bench (AA)Artificial Analysis26.2%12.5%†
CyberGym end-to-end (AA)Artificial Analysis78.6%59.5%†
MKA (AA)Artificial Analysis37.3%–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub–74.0%†
FrontierMath Tier 4Epoch AI Benchmarking Hub–46.3%†
OTIS Mock AIMEEpoch AI Benchmarking Hub–99.2%†
ProofBenchVals AI70.0%55.0%
LiveBench MathematicsLiveBench–96.0%†
Arena MathsArena (LMArena)14841506†
Knowledge
LiveBench 2.0 Data AnalysisLiveBench72.3%†79.6%†
LiveBench ResearchLiveBench84.8%†82.5%†
LiveBench 1.0 Data AnalysisLiveBench–79.6%†
AA-Omniscience v1.1Artificial Analysis-4.522.4†
PRBench FinanceScale AI SEAL–59.5%
PRBench LegalScale AI SEAL–61.6%
Excel Modeling BenchmarkVals AI62.9%67.4%†
Excel Modeling Benchmark v1.1Vals AI67.3%–
MedCodeVals AI45.0%–
MedScribeVals AI88.3%–
Public Benefits Bench v1.1Vals AI68.9%–
Access to Justice BenchmarkVals AI70.0%72.2%†
Financial Coaching BenchVals AI75.4%70.3%†
Immigration BenchVals AI67.5%64.9%†
DrugDiscoveryBenchScale AI SEAL–62.2%†
GDP.pdfEpoch AI Benchmarking Hub–27.6%
AA-Omniscience v1.1: accuracyArtificial Analysis34.8%38.0%†
AA-Omniscience v1.1: non-hallucinationArtificial Analysis39.8%74.8%†
Instruction following
LiveBench LanguageLiveBench–82.8%†
LiveBench Instruction FollowingLiveBench–78.0%†
Arena Instruction FollowingArena (LMArena)14721485†
Human preference
Arena TextArena (LMArena)14801497†
EQ-Bench Creative Writing v3EQ-Bench–1907
Arena Creative WritingArena (LMArena)14481468†
Arena Multi-turnArena (LMArena)14621493†
Arena Longer QueriesArena (LMArena)14961498†
Multimodal
Arena VisionArena (LMArena)12661310†
LiveBench DesignLiveBench74.3%†74.8%†
MMMU-ProArtificial Analysis–82.0%†
SAGEVals AI84.5%–
Humanity's Sixth SenseScale AI SEAL25.2%†37.4%†
GeoBenchGeoBench–86.0%
VIALS-AAHandshake AI43.5%–
Code2Video-AAArtificial Analysis10521019†
AA-3DArtificial Analysis9831032†
Long context
AA-LCR v1.1Artificial Analysis86.3%83.0%†
Arena DocumentArena (LMArena)–1474†
MLCRArtificial Analysis12.8%13.9%†
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub–158.2
LiveBench 2.0LiveBench66.1%†68.6%†
LiveBench 1.0LiveBench–81.6%†
AA Intelligence Index v5.0.1Artificial Analysis29.936.6†
AA Intelligence Index v4.3.2Artificial Analysis46.348.1†
AA Progress Index v1Artificial Analysis321.2324.5†
Vals IndexVals AI57.8–
Vals Finance IndexVals AI57.9%–
Vals RSI IndexVals AI–19.6%†
Vals Router IndexVals AI57.8%–
Vals Legal IndexVals AI72.1%–
AA Cyber IndexArtificial Analysis56.144.3†
Safety & honesty
DistressBenchScale AI SEAL–84.9%
Metadata
AA Openness IndexArtificial Analysis52.8–

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Muse Spark 1.3: BenchLeader Index 60.2 → 61.6.
  • MiMo-V2.6-Pro: 1 score revision.
  • MiMo-V2.6-Pro: 1 score revision.
  • Muse Spark 1.3: 1 score revision.
  • Muse Spark 1.3: BenchLeader Index 59.2 → 60.2.
  • Muse Spark 1.3: 2 new results (SimpleBench, GeoBench).
  • Muse Spark 1.3: BenchLeader Index 57.7 → 59.2.
  • MiMo-V2.6-Pro: 6 score revisions.
  • Muse Spark 1.3: 4 score revisions.
  • MiMo-V2.6-Pro: 1 score revision.

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

MiMo-V2.6-Pro vs Muse Spark 1.3: questions

Is MiMo-V2.6-Pro better than Muse Spark 1.3?
MiMo-V2.6-Pro and Muse Spark 1.3 are neck and neck on quality (57.5 vs 61.6, within the ±9.1 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
MiMo-V2.6-Pro vs Muse Spark 1.3: which is cheaper, and will that change?
MiMo-V2.6-Pro is cheaper today: $0.544 against $2.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Which is faster, MiMo-V2.6-Pro or Muse Spark 1.3?
Muse Spark 1.3 streams ~1.5× faster (163 vs 54 output tokens per second, Artificial Analysis 7-day median).
Is MiMo-V2.6-Pro or Muse Spark 1.3 better for agents and tool use?
Muse Spark 1.3 scores higher in agentic tasks (61 vs 57 on the category index, where 50 is average). On APEX-Agents, the top shared agents & tools benchmark, it is 54.4% for MiMo-V2.6-Pro against 58.6% for Muse Spark 1.3.
Which is better for long documents?
Both accept 1,048,576 tokens of context.
More questions (3)
Which has lower latency?
Muse Spark 1.3 answers first: 2.6 s against 42.3 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on MiMo-V2.6-Pro vs Muse Spark 1.3?
Across the 2 benchmarks both have, MiMo-V2.6-Pro leads 1, Muse Spark 1.3 leads 1. Every overall index that rates both puts Muse Spark 1.3 ahead.
When was MiMo-V2.6-Pro released?
MiMo-V2.6-Pro was released on 21 Sep 2026.