BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

GLM 5.3 vs MiMo-V2.6-Pro

Full profiles: GLM 5.3 (Zhipu AI) and MiMo-V2.6-Pro (Xiaomi): every benchmark, price and rank.

Updated · Data from 12 sources

As of 9 Nov 2026, GLM 5.3 and MiMo-V2.6-Pro are neck and neck on quality (58.9 vs 57.1 on the BenchLeader Index, inside the ±4.9 margin), and MiMo-V2.6-Pro is 4.0× cheaper at $0.544 vs $2.15 per 1M tokens blended. Choose GLM 5.3 for knowledge (54 vs 42) and reasoning (55 vs 50), where it also streams faster; choose MiMo-V2.6-Pro for fast first answers (25.7 s vs 80.5 s) and maths (66 vs 60) and long context (68 vs 65).

Key facts (as of 9 Nov 2026)

  • GLM 5.3 scores 58.9 on the BenchLeader Index and MiMo-V2.6-Pro scores 57.1 (BenchLeader, 9 Nov 2026).
  • GLM 5.3 costs $2.15 per 1M tokens blended (list price, 9 Nov 2026).
  • MiMo-V2.6-Pro costs $0.544 per 1M tokens blended (list price, 9 Nov 2026).
  • GLM 5.3 streams faster than MiMo-V2.6-Pro (Artificial Analysis, 7-day median, 9 Nov 2026).
  • Of the 61 benchmarks both models have, MiMo-V2.6-Pro leads 31 (BenchLeader, 9 Nov 2026).
  • On Arena WebDev, GLM 5.3 is rated 1623 and MiMo-V2.6-Pro 1624 (LMArena, as published by 9 Nov 2026).
GLM 5.3 benchmarks max Zhipu AI MiMo-V2.6-Pro benchmarks XiaomiAdd a third model

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
CodingEither60 vs 59 category score; Arena WebDev 1623 vs 1624Either
Agents & tool useEither58 vs 58 category score; Arena Agent 1 vs 2.3Either
MathsMiMo-V2.6-Pro66 vs 60 category score; ProofBench 70.0% vs 49.0%Medium
ReasoningGLM 5.355 vs 50 category score; HLE-Diamond (AA) 17.9% vs 15.2%Medium
KnowledgeGLM 5.354 vs 42 category score; AA-Omniscience v1.1 6.0 vs -4.5High
Long documentsMiMo-V2.6-Pro68 vs 65 category score; AA-LCR v1.1 86.3% vs 79.7%Medium
Chat quality (human preference)Either67 vs 67 on Arena votesEither
Lowest costMiMo-V2.6-Pro$0.548 vs $2.15 per 1M blended; agentic coding session $0.030 vs $0.102–
Fastest first answerMiMo-V2.6-Pro25.7 s (MiMo-V2.6-Pro) vs 80.5 s (GLM 5.3 (max)) 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

  • GLM 5.3 (max) and MiMo-V2.6-Pro are neck and neck on quality (58.9 vs 57.1, within the ±4.9 margin).
  • GLM 5.3 (max) is stronger in coding, human preference, knowledge, reasoning.
  • MiMo-V2.6-Pro is stronger in agents & tools, long context, maths, composite, multimodal.
  • MiMo-V2.6-Pro is 4.0× cheaper ($0.544 vs $2.15 per 1M blended).
  • GLM 5.3 (max) streams 1.3× faster (67 vs 51 tokens per second).

At a glance

SpecGLM 5.3MiMo-V2.6-Pro
DeveloperZhipu AIXiaomi
Released18 Aug 202621 Sep 2026
Context window (tokens)1,000,0001,048,576
Compared atGLM 5.3 (max)MiMo-V2.6-Pro
Open weightsYesYes
Price in / out / cached input per 1M$1.40 / $4.40 / $0.260$0.440 / $0.870 / $0.0036

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.

CategoryGLM 5.3 (max)MiMo-V2.6-ProLeader
Reasoning55.050.2GLM 5.3
Coding59.959.1level
Agents & tools57.858.4level
Maths60.365.6MiMo-V2.6-Pro
Knowledge53.742.3GLM 5.3
Human preference67.467.1level
Multimodal–60.2not comparable
Long context64.568.0MiMo-V2.6-Pro
Composite–51.9not 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 avg75ReasoningCodingAgents & toolsMathsKnowledgeLong contextHumanpreferenceGLM 5.3, Reasoning: 55.0GLM 5.3, Coding: 59.9GLM 5.3, Agents & tools: 57.8GLM 5.3, Maths: 60.3GLM 5.3, Knowledge: 53.7GLM 5.3, Long context: 64.5GLM 5.3, Human preference: 67.4MiMo-V2.6-Pro, Reasoning: 50.2MiMo-V2.6-Pro, Coding: 59.1MiMo-V2.6-Pro, Agents & tools: 58.4MiMo-V2.6-Pro, Maths: 65.6MiMo-V2.6-Pro, Knowledge: 42.3MiMo-V2.6-Pro, Long context: 68.0MiMo-V2.6-Pro, Human preference: 67.1

Not drawn, measured for fewer than two of these models: Instruction following, Multimodal.

ModelIndex

Where sources disagree

Across the 61 benchmarks both have, GLM 5.3 leads 30, MiMo-V2.6-Pro leads 31. Every overall index that rates both puts GLM 5.3 ahead.

Overall indexGLM 5.3 (max)MiMo-V2.6-ProLeader
BenchLeader Index58.957.1neck and neck
AA Intelligence Index v5.031.629.9GLM 5.3
Vals Index59.957.8GLM 5.3
Arena Text14831480GLM 5.3

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.

MeasureGLM 5.3 (max)MiMo-V2.6-Pro
Blended price per 1M tokens$2.15$0.548
Input / output per 1M$1.40 / $4.40$0.440 / $0.870
Cached input per 1M$0.260$0.0036
Price per task: chat reply$0.0019$0.0004
Price per task: summarize a 30-page report$0.019$0.0058
Price per task: code edit$0.015$0.0039
Price per task: agentic coding session$0.102$0.030
Price per task: structured extraction$0.0037$0.0011
Agentic coding session, input mostly cached$0.050$0.010
Cost per benchmark task (AA Index)$5.01$0.257
Output tokens per task (AA Index)119k121k
Of which reasoning85k86k
Time per task (AA Index)30 min39 min
Output speed67 tok/s51 tok/s
Time to first answer80.5 s25.7 s
Time to a complete answer155.4 s100.9 s

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $5.01 on GLM 5.3 (max) and $0.257 on MiMo-V2.6-Pro, and MiMo-V2.6-Pro wrote 1.0× as many output tokens for the same work.

Each model is compared at its best-ranked reasoning-effort setting. GLM 5.3 (max) scores 58.9 and takes 80 s to its first answer token; GLM 5.3 (low) scores 47.4 in 12 s. MiMo-V2.6-Pro scores 57.1 and takes 26 s to its first answer token. 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

131 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.

MetricPublisherGLM 5.3 (max)MiMo-V2.6-Pro
BenchLeader IndexBenchLeader58.957.1
Agents & tools scoreBenchLeader57.858.4
Coding scoreBenchLeader59.959.1
Human preference scoreBenchLeader67.467.1
Knowledge scoreBenchLeader53.742.3
Long context scoreBenchLeader64.568.0
Maths scoreBenchLeader60.365.6
Reasoning scoreBenchLeader55.050.2
Composite scoreBenchLeader–51.9
Multimodal scoreBenchLeader–60.2
Blended price $/MBenchLeader$2.15$0.548
Output speedBenchLeader67 tok/s51 tok/s
Time to first answerBenchLeader80.5 s25.7 s
Context windowBenchLeader1,000,000 tokens1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub90.9%–
Arena Hard PromptsArena (LMArena)15071504
LiveBench 2.0 ReasoningLiveBench–79.6%
LiveBench 1.0 ReasoningLiveBench85.8%†–
GPQA Diamond (AA)Artificial Analysis91.7%–
Humanity's Last Exam (AA)Artificial Analysis42.3%49.4%
HLE-Diamond (AA)Artificial Analysis17.9%15.2%
GPQA Diamond (Vals)Vals AI88.1%–
CritPtArtificial Analysis19.1%26.6%
CritPt v2 hard subset (AA)Artificial Analysis33.5%34.0%
MysteryMechanismVals AI23.0%15.3%
Chess PuzzlesEpoch AI Benchmarking Hub37.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub33.0%–
LMCAEpoch AI Benchmarking Hub55.5%†–
DTBenchEpoch AI Benchmarking Hub87.7%†–
BTF-3Epoch AI Benchmarking Hub14.9%†–
Coding
SciCodeSciCode59.0%60.9%
WeirdMLWeirdML75.4%–
FrontierCodeCognition40.1%–
Arena CodingArena (LMArena)15221538
Arena WebDevArena (LMArena)16231624
LiveBench 2.0 CodingLiveBench–64.5%
LiveBench 1.0 CodingLiveBench79.0%†–
LiveCodeBenchVals AI80.5%–
IOIVals AI68.4%39.3%
Show all 131 benchmarks
MetricPublisherGLM 5.3 (max)MiMo-V2.6-Pro
Coding
SWE-bench (Vals)Vals AI95.4%–
Code MigrationVals AI44.2%43.0%
Code Migration v1.1Vals AI55.1%48.8%
ProgramBenchVals AI1.5%0.5%
SWE-bench MultimodalVals AI–41.3%
Vibe Code Bench 1-100Vals AI20.0%–
Vibe Code Bench v1.1Vals AI78.1%85.2%
DeepSWE v1.1Epoch AI Benchmarking Hub69.0%–
CursorBenchCursor42.6%–
ALE-BenchEpoch AI Benchmarking Hub1317.4†1158.1
FrontierSWEEpoch AI Benchmarking Hub30.2%–
Terminal-Bench 4.0 (AA)Artificial Analysis41.9%34.9%
Terminal-Bench 2.1 (AA)Artificial Analysis83.9%–
Startrise FrontendStartrise73.2%†–
GameBench 2 OverallDuelLab GameBench 264.1†–
GameBench 2 by reasoning settingDuelLab GameBench 272.149.6†
Design Arena WebsiteDesign Arena1286†1305
FrontierCode 1.1 (AA)Artificial Analysis40.1%45.4%
FrontierCode 2.0 (AA)Artificial Analysis21.6%24.0%
Agents & tools
Terminal-BenchTerminal-Bench41.2%–
APEX-AgentsMercor56.6%†54.4%
Arena AgentArena (LMArena)12.3
LiveBench Long AgenticLiveBench–36.4%
LiveBench BrowserLiveBench–77.3%
LiveBench Agentic CodingLiveBench60.9%†–
Terminal-Bench 2.1 (Vals)Vals AI71.5%67.8%
MCP AtlasScale AI SEAL84.2%†–
HiL-BenchScale AI SEAL58.3%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis57.5%59.1%
GDPval-AA v2.1Artificial Analysis16571689
τ³-Banking (AA)Artificial Analysis50.3%–
ITBench SRE (AA)Artificial Analysis46.1%–
Analyst Agent (AA)Artificial Analysis38.8%37.5%
CyberBenchVals AI72.1%72.9%
Finance Agent v2Vals AI55.8%57.3%
Finance Agent v2.1Vals AI65.0%64.3%
Harvey's Legal Agent Benchmark v1.1Vals AI14.6%13.8%
Harvey's Legal Agent Benchmark v1Vals AI8.3%10.8%
Legal Research BenchVals AI49.0%47.1%
SkillsBenchVals AI47.5%–
SREBenchVals AI9.2%3.0%
Tax Agent BenchVals AI43.3%32.4%
Terminal-Bench 4.0 (Vals)Vals AI38.9%31.3%
Terminal-Bench ScienceVals AI5.7%2.9%
Vending-Bench 2Andon Labs8163.6†–
AutomationBenchArtificial Analysis62.2%58.6%
GDP.pdf (AA)Artificial Analysis11.2%19.2%
Harvey LAB-AA v1.1Harvey0.3%–
EnterpriseOps-GymArtificial Analysis36.4%–
AA-Briefcase v1.1Artificial Analysis15031512
GDPxGDPx by SemiAnalysis2.7%–
Design Arena AgenticDesign Arena1230†1266
Terminal-Bench-Science 0.1 (AA)Artificial Analysis9.5%5.7%
CWE-Bench (AA)Artificial Analysis55.8%63.3%
DeepSec-Bench (AA)Artificial Analysis24.3%26.2%
CyberGym end-to-end (AA)Artificial Analysis29.0%78.6%
FACTS SearchGoogle87.7%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub68.8%–
FrontierMath Tier 4Epoch AI Benchmarking Hub29.3%–
OTIS Mock AIMEEpoch AI Benchmarking Hub91.1%–
ProofBenchVals AI49.0%70.0%
LiveBench MathematicsLiveBench87.9%†–
Arena MathsArena (LMArena)15021482
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub41.0%–
LiveBench 2.0 Data AnalysisLiveBench–72.3%
LiveBench ResearchLiveBench–84.8%
LiveBench 1.0 Data AnalysisLiveBench70.2%†–
AA-Omniscience v1.1Artificial Analysis6.0-4.5
MMLU-ProVals AI86.8%–
LegalBenchVals AI84.8%–
TaxEvalVals AI72.4%–
Excel Modeling BenchmarkVals AI56.3%62.9%
Excel Modeling Benchmark v1.1Vals AI69.1%67.3%
MedCodeVals AI42.9%45.0%
MedScribeVals AI88.8%88.3%
Public Benefits Bench v1.1Vals AI68.5%68.9%
AA-Omniscience v1.1: accuracyArtificial Analysis32.3%34.8%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis61.1%39.8%
HealthBench Professional (AA)OpenAI52.8%–
Instruction following
LiveBench LanguageLiveBench79.9%†–
LiveBench Instruction FollowingLiveBench69.3%†–
Arena Instruction FollowingArena (LMArena)14801472
Human preference
Arena TextArena (LMArena)14831480
Arena Creative WritingArena (LMArena)14571448
Arena Multi-turnArena (LMArena)14821463
Arena Longer QueriesArena (LMArena)14891497
Multimodal
Arena VisionArena (LMArena)–1266
LiveBench DesignLiveBench–74.3%
SAGEVals AI–84.5%
Code2Video-AAArtificial Analysis10341052
AA-3DArtificial Analysis794983
Long context
AA-LCR v1.1Artificial Analysis79.7%86.3%
MLCRArtificial Analysis17.2%12.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub156.9†–
LiveBench 2.0LiveBench–66.1%
LiveBench 1.0LiveBench76.1%†–
AA Intelligence Index v5.0Artificial Analysis31.629.9
AA Intelligence Index v4.3.2Artificial Analysis44.846.3
AA Progress Index v1Artificial Analysis322.7321.2
Vals IndexVals AI59.957.8
Vals Finance IndexVals AI61.4%57.9%
Vals RSI IndexVals AI22.9%–
Vals Router IndexVals AI59.7%†57.8%
AA Cyber IndexArtificial Analysis36.456.1
Safety & honesty
FORTRESSScale AI SEAL28.2%–
Metadata
AA Openness IndexArtificial Analysis33.352.8

Data as of 2026-11-09. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • GLM 5.3 (max): 4 score revisions.
  • MiMo-V2.6-Pro: 6 score revisions.
  • MiMo-V2.6-Pro: 1 score revision.
  • GLM 5.3 (max): 1 score revision; cached input price moved from $0.30/M to $0.26/M (13% down).
  • GLM 5.3 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
  • GLM 5.3 (max): cached input price moved from $0.30/M to $0.26/M (13% down).
  • GLM 5.3 (max): cached input price moved from $0.26/M to $0.30/M (15% up).
  • MiMo-V2.6-Pro: 1 score revision.
  • GLM 5.3 (max): BenchLeader Index 61.3 → 58.9.
  • MiMo-V2.6-Pro: BenchLeader Index 61.1 → 57.2.

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

GLM 5.3 vs MiMo-V2.6-Pro: questions

Is GLM 5.3 better than MiMo-V2.6-Pro?
GLM 5.3 (max) and MiMo-V2.6-Pro are neck and neck on quality (58.9 vs 57.1, within the ±4.9 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 9 Nov 2026, so check the category scores for your use.
GLM 5.3 vs MiMo-V2.6-Pro: which is cheaper, and will that change?
MiMo-V2.6-Pro is cheaper today: $0.544 against $2.15 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is GLM 5.3 or MiMo-V2.6-Pro better for coding?
They are level in coding (60 vs 59 on the category index). On Arena WebDev, the top shared coding benchmark, it is 1623 for GLM 5.3 against 1624 for MiMo-V2.6-Pro.
Which is faster, GLM 5.3 or MiMo-V2.6-Pro?
GLM 5.3 streams faster (67 vs 51 output tokens per second, Artificial Analysis 7-day median).
Is GLM 5.3 or MiMo-V2.6-Pro better for agents and tool use?
They are level in agentic tasks (58 vs 58 on the category index). On Arena Agent, the top shared agents & tools benchmark, it is 1 for GLM 5.3 against 2.3 for MiMo-V2.6-Pro.
More questions (4)
Which is better for long documents?
MiMo-V2.6-Pro scores higher on long-context benchmarks (68 vs 65). Both take about 1M tokens: GLM 5.3 1,000,000, MiMo-V2.6-Pro 1,048,576.
Which has lower latency?
MiMo-V2.6-Pro answers first: 25.7 s against 80.5 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 GLM 5.3 vs MiMo-V2.6-Pro?
Across the 61 benchmarks both have, GLM 5.3 leads 30, MiMo-V2.6-Pro leads 31. Every overall index that rates both puts GLM 5.3 ahead.
When was MiMo-V2.6-Pro released?
MiMo-V2.6-Pro was released on 21 Sep 2026.