BenchLeader

GPT-5.4 vs MiMo-V2.6-Pro

Full profiles: GPT-5.4 (OpenAI) and MiMo-V2.6-Pro (Xiaomi): every benchmark, price and rank.

Updated · Data from 13 sources

As of 16 Nov 2026, GPT-5.4 is ahead of MiMo-V2.6-Pro on quality (63.1 vs 57.5 on the BenchLeader Index), but MiMo-V2.6-Pro is 10× cheaper at $0.544 vs $5.63 per 1M tokens blended. Choose GPT-5.4 for knowledge (59 vs 42), reasoning (66 vs 50), and coding (65 vs 59), where it also streams faster; choose MiMo-V2.6-Pro for fast first answers (42.3 s vs 144.9 s) and long context (68 vs 66).

Key facts (as of 16 Nov 2026)

  • GPT-5.4 scores 63.1 on the BenchLeader Index and MiMo-V2.6-Pro scores 57.5 (BenchLeader, 16 Nov 2026).
  • GPT-5.4 costs $5.63 per 1M tokens blended (list price, 16 Nov 2026).
  • MiMo-V2.6-Pro costs $0.544 per 1M tokens blended (list price, 16 Nov 2026).
  • GPT-5.4 streams faster than MiMo-V2.6-Pro (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 13 benchmarks both models have, MiMo-V2.6-Pro leads 13 (BenchLeader, 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
CodingGPT-5.465 vs 59 category score; SciCode 56.6% vs 60.9%Medium
Agents & tool useGPT-5.463 vs 57 category score; GDPval-AA v2.1 1256 vs 1692Medium
MathsEither65 vs 66 category score; ProofBench 56.0% vs 70.0%Either
ReasoningGPT-5.466 vs 50 category score; Humanity's Last Exam (AA) 43.7% vs 49.4%High
KnowledgeGPT-5.459 vs 42 category score; MedCode 41.3% vs 45.0%High
Long documentsMiMo-V2.6-Pro68 vs 66 category score; AA-LCR v1.1 86.3% vs 82.0%Medium
Images and other mediaEither61 vs 60 category scoreEither
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 $5.63 per 1M blended; agentic coding session $0.030 vs $0.210–
Fastest first answerMiMo-V2.6-Pro42.3 s (MiMo-V2.6-Pro) vs 144.9 s (GPT-5.4 (xhigh)) 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

  • GPT-5.4 (xhigh) leads on quality: 63.1 vs 57.5.
  • GPT-5.4 (xhigh) is stronger in agents & tools, coding, knowledge, multimodal, reasoning.
  • MiMo-V2.6-Pro is stronger in long context, maths.
  • MiMo-V2.6-Pro is 10× cheaper ($0.544 vs $5.63 per 1M blended).
  • GPT-5.4 (xhigh) streams 1.3× faster (73 vs 54 tokens per second).

At a glance

SpecGPT-5.4MiMo-V2.6-Pro
DeveloperOpenAIXiaomi
Released5 Mar 202621 Sep 2026
Context window (tokens)1,050,0001,048,576
Compared atGPT-5.4 (xhigh)MiMo-V2.6-Pro
Open weightsNoYes
Price in / out / cached input per 1M$2.50 / $15.00 / $0.250$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.

CategoryGPT-5.4 (xhigh)MiMo-V2.6-ProLeader
Reasoning65.749.9GPT-5.4
Coding64.758.8GPT-5.4
Agents & tools63.057.1GPT-5.4
Maths64.865.5level
Knowledge58.642.3GPT-5.4
Instruction following72.4–not comparable
Human preference–67.0not comparable
Multimodal60.860.2level
Long context65.767.9MiMo-V2.6-Pro

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 contextMultimodalGPT-5.4, Reasoning: 65.7GPT-5.4, Coding: 64.7GPT-5.4, Agents & tools: 63.0GPT-5.4, Maths: 64.8GPT-5.4, Knowledge: 58.6GPT-5.4, Long context: 65.7GPT-5.4, Multimodal: 60.8MiMo-V2.6-Pro, Reasoning: 49.9MiMo-V2.6-Pro, Coding: 58.8MiMo-V2.6-Pro, Agents & tools: 57.1MiMo-V2.6-Pro, Maths: 65.5MiMo-V2.6-Pro, Knowledge: 42.3MiMo-V2.6-Pro, Long context: 67.9MiMo-V2.6-Pro, Multimodal: 60.2

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

Model, click to add or removeIndex

Where sources disagree

Across the 13 benchmarks both have, GPT-5.4 leads 0, MiMo-V2.6-Pro leads 13. Independent sources split: GPT-5.4 leads on BenchLeader Index; MiMo-V2.6-Pro leads on Arena Text.

Overall indexGPT-5.4 (xhigh)MiMo-V2.6-ProLeader
BenchLeader Index63.157.5GPT-5.4
Arena Text1475†1480MiMo-V2.6-Pro

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.

MeasureGPT-5.4 (xhigh)MiMo-V2.6-Pro
Blended price per 1M tokens$5.63$0.548
Input / output per 1M$2.50 / $15.00$0.440 / $0.870
Cached input per 1M$0.250$0.0036
Price per task: chat reply$0.0055$0.0004
Price per task: summarize a 30-page report$0.039$0.0058
Price per task: code edit$0.037$0.0039
Price per task: agentic coding session$0.210$0.030
Price per task: structured extraction$0.0080$0.0011
Agentic coding session, input mostly cached$0.109$0.010
Cost per benchmark task (AA Index)–$0.257
Output tokens per task (AA Index)–121k
Of which reasoning–86k
Time per task (AA Index)–38 min
Output speed73 tok/s54 tok/s
Time to first answer144.9 s42.3 s
Time to a complete answer, last measured 13 Nov 2026193.1 s95.7 s

Each model is compared at its best-ranked reasoning-effort setting. GPT-5.4 (xhigh) scores 63.1 and takes 145 s to its first answer token; GPT-5.4 (low) scores 58.8 in 3 s, GPT-5.4 (high) scores 58.6 in 2 s. MiMo-V2.6-Pro scores 57.5 and takes 42 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

181 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.

MetricPublisherGPT-5.4 (xhigh)MiMo-V2.6-Pro
BenchLeader IndexBenchLeader63.157.5
Agents & tools scoreBenchLeader63.057.1
Coding scoreBenchLeader64.758.8
Instruction following scoreBenchLeader72.4–
Knowledge scoreBenchLeader58.642.3
Long context scoreBenchLeader65.767.9
Maths scoreBenchLeader64.865.5
Multimodal scoreBenchLeader60.860.2
Reasoning scoreBenchLeader65.749.9
Human preference scoreBenchLeader–67.0
Blended price $/MBenchLeader$5.63$0.548
Output speedBenchLeader73 tok/s54 tok/s
Time to first answerBenchLeader144.9 s42.3 s
Context windowBenchLeader1,050,000 tokens1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub93.3%–
Humanity's Last ExamScale AI / CAIS36.2%–
Arena Hard PromptsArena (LMArena)1487†1504
LiveBench 2.0 ReasoningLiveBench–79.6%†
LiveBench 1.0 ReasoningLiveBench88.1%–
GPQA Diamond (AA)Artificial Analysis92.0%–
Humanity's Last Exam (AA)Artificial Analysis43.7%49.4%
HLE-Diamond (AA)Artificial Analysis–15.2%
GPQA Diamond (Vals)Vals AI91.7%–
Kagi LLM BenchmarkKagi LLM Benchmark63.8%†–
ARC-AGI-1ARC Prize93.7%–
ARC-AGI-2ARC Prize74.0%–
ARC-AGI-3ARC Prize0.2%†–
CritPtArtificial Analysis23.4%26.6%
CritPt v2 hard subset (AA)Artificial Analysis–34.0%
MysteryMechanismVals AI–15.3%
Arena ExpertArena (LMArena)1491†1526
Chess PuzzlesEpoch AI Benchmarking Hub48.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub37.0%–
CL-benchEpoch AI Benchmarking Hub27.9%–
CL-bench LifeEpoch AI Benchmarking Hub21.7%–
LMCAEpoch AI Benchmarking Hub52.0%–
DTBenchEpoch AI Benchmarking Hub94.4%–
ForecastBenchForecastBench63.0%†–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub76.9%†–
Show all 181 benchmarks
MetricPublisherGPT-5.4 (xhigh)MiMo-V2.6-Pro
Coding
SciCodeSciCode56.6%60.9%
WeirdMLWeirdML77.7%–
GSO-BenchGSO-Bench31.4%–
Arena CodingArena (LMArena)1511†1538
Arena WebDevArena (LMArena)1400†1625
LiveBench 2.0 CodingLiveBench–64.5%†
LiveBench 1.0 CodingLiveBench77.5%–
SciCode (AA)Artificial Analysis–60.9%
LiveCodeBenchVals AI84.1%–
IOIVals AI–39.3%
SWE-bench (Vals)Vals AI78.2%–
SWE-Bench ProScale AI SEAL59.1%–
Code MigrationVals AI35.0%43.0%
Code Migration v1.1Vals AI–48.8%
ProgramBenchVals AI0.0%0.5%
SWE-bench MultimodalVals AI–41.3%
Vibe Code Bench v1.1Vals AI67.4%85.2%
Vibe Code Bench 1-100 v1.1Vals AI–26.4%
SWE-Bench Pro (private)Scale AI SEAL43.4%–
SWE Atlas v1.0: Codebase QnAScale AI SEAL40.8%–
SWE Atlas v1.0: RefactoringScale AI SEAL44.3%–
SWE Atlas v1.0: Test WritingScale AI SEAL44.4%–
MirrorCodeEpoch AI Benchmarking Hub23.3%–
DeepSWE v1.1Epoch AI Benchmarking Hub51.8%–
ALE-BenchEpoch AI Benchmarking Hub1607†1158.1
AlgoTuneAlgoTune1.9†–
Terminal-Bench 4.0 (AA)Artificial Analysis–34.9%
Terminal-Bench 2.1 (AA)Artificial Analysis78.3%–
GameBench 2 OverallDuelLab GameBench 253.4†–
GameBench 2 by reasoning settingDuelLab GameBench 262.149.8†
Design Arena WebsiteDesign Arena1190†1311
FrontierCode 1.1 (AA)Artificial Analysis–45.4%
FrontierCode 2.0 (AA)Artificial Analysis–24.0%
Agents & tools
Terminal-BenchTerminal-Bench81.8%†27.0%†
APEX-AgentsMercor52.4%†54.4%
Arena AgentArena (LMArena)–2.4
LiveBench Long AgenticLiveBench–36.4%†
LiveBench BrowserLiveBench–77.3%†
LiveBench Agentic CodingLiveBench53.8%–
Terminal-Bench HardArtificial Analysis57.6%–
τ²-Bench Telecom (AA)Artificial Analysis87.1%–
Terminal-Bench 2.1 (Vals)Vals AI–67.8%
MCP AtlasScale AI SEAL70.6%–
HiL-BenchScale AI SEAL9.7%†–
τ³-bench Bankingτ²-bench39.4%–
GDPval-AA v2.1, normalised scoreArtificial Analysis–59.1%
GDPval-AA v2.1Artificial Analysis12561692
τ³-Banking (AA)Artificial Analysis39.6%–
Analyst Agent (AA)Artificial Analysis–37.5%
APEX-Agents (AA)Artificial Analysis33.3%–
CyberBenchVals AI–72.9%
Finance Agent v2Vals AI–57.3%
Finance Agent v2.1Vals AI–64.3%
Harvey's Legal Agent Benchmark v1.1Vals AI–13.8%
Harvey's Legal Agent Benchmark v1Vals AI0.0%10.8%
Legal Research BenchVals AI–47.1%
SkillsBenchVals AI51.7%–
SREBenchVals AI–3.0%
Tax Agent BenchVals AI–32.4%
Terminal-Bench 2.0 (Vals)Vals AI58.4%†–
Terminal-Bench 4.0 (Vals)Vals AI–31.3%
Terminal-Bench ScienceVals AI–2.9%
Arena SearchArena (LMArena)1196†–
Arena Agent Task CompletionArena (LMArena)–5
DeepResearch BenchDeepResearch Bench35.1%†–
PostTrainBenchEpoch AI Benchmarking Hub17.8%†–
METR Time HorizonsMETR74.3%–
GBAEvalEpoch AI Benchmarking Hub45.1%†–
Vending-Bench 2Andon Labs6144.2†–
AutomationBenchArtificial Analysis–58.6%
GDP.pdf (AA)Artificial Analysis–19.2%
AA-Briefcase v1.1Artificial Analysis–1512
Design Arena AgenticDesign Arena1042†1271
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–5.7%
CWE-Bench (AA)Artificial Analysis–63.3%
DeepSec-Bench (AA)Artificial Analysis–26.2%
CyberGym end-to-end (AA)Artificial Analysis–78.6%
MKA (AA)Artificial Analysis–37.3%
DeepSearchQAGoogle63.7%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub78.6%–
FrontierMath Tier 4Epoch AI Benchmarking Hub49.0%–
OTIS Mock AIMEEpoch AI Benchmarking Hub95.3%–
ProofBenchVals AI56.0%70.0%
LiveBench MathematicsLiveBench94.2%–
AIME (Vals)Vals AI96.7%–
AIME 2026MathArena99.2%–
HMMT February 2026MathArena97.7%–
USAMO 2026MathArena95.2%–
MathArena ApexMathArena54.2%–
Arena MathsArena (LMArena)1459†1484
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub45.1%–
LiveBench 2.0 Data AnalysisLiveBench–72.3%†
LiveBench ResearchLiveBench–84.8%†
LiveBench 1.0 Data AnalysisLiveBench79.3%–
AA-Omniscience v1.1Artificial Analysis–-4.5
MMLU-ProVals AI87.5%–
LegalBenchVals AI86.0%–
CorpFinVals AI65.3%–
TaxEvalVals AI74.0%–
MedQAVals AI96.1%–
PRBench FinanceScale AI SEAL45.6%†–
PRBench LegalScale AI SEAL44.4%†–
MultiNRCScale AI SEAL58.3%†–
CaseLaw v2Vals AI63.8%–
Excel Modeling BenchmarkVals AI–62.9%
Excel Modeling Benchmark v1.1Vals AI–67.3%
MedCodeVals AI41.3%45.0%
MedScribeVals AI77.5%88.3%
Public Benefits Bench v1.1Vals AI–68.9%
Access to Justice BenchmarkVals AI–70.0%
Financial Coaching BenchVals AI–75.4%
Immigration BenchVals AI–67.5%
EBR-benchEpoch AI Benchmarking Hub25.4%–
AA-Omniscience v1.1: accuracyArtificial Analysis–34.8%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–39.8%
Instruction following
LiveBench LanguageLiveBench82.6%–
LiveBench Instruction FollowingLiveBench70.2%–
IFBenchArtificial Analysis74.0%–
Arena Instruction FollowingArena (LMArena)1460†1472
Human preference
Arena TextArena (LMArena)1465†1480
EQ-Bench 4EQ-Bench1258†–
EQ-Bench Creative Writing v3EQ-Bench1841†–
Arena Creative WritingArena (LMArena)1433†1448
Arena Multi-turnArena (LMArena)1480†1462
Arena Longer QueriesArena (LMArena)1477†1496
Multimodal
Arena VisionArena (LMArena)1292†1266
LiveBench DesignLiveBench–74.3%†
MMMU-ProArtificial Analysis78.4%–
VISTAScale AI SEAL50.9%†–
MMMU-Pro (Vals)Vals AI87.5%–
MortgageTaxVals AI68.3%–
SAGEVals AI–84.5%
Humanity's Sixth SenseScale AI SEAL–25.2%†
VTBScale AI SEAL29.2%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub27.1%†–
Furniture AssemblyEpoch AI Benchmarking Hub37.5%–
VIALS-AAHandshake AI–43.5%
Code2Video-AAArtificial Analysis–1052
AA-3DArtificial Analysis–983
Long context
AA-LCR v1.1Artificial Analysis82.0%86.3%
Arena DocumentArena (LMArena)1471†–
MLCRArtificial Analysis–12.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub157.8†–
LiveBench 2.0LiveBench–66.1%†
LiveBench 1.0LiveBench78.0%–
AA Intelligence Index v5.0.1Artificial Analysis–29.9
AA Intelligence Index v4.3.2Artificial Analysis39.046.3
AA Progress Index v1Artificial Analysis311.6321.2
Vals IndexVals AI–57.8
Vals Finance IndexVals AI–57.9%
Vals RSI IndexVals AI15.5%–
Vals Router IndexVals AI–57.8%
Vals Legal IndexVals AI–72.1%
AA Cyber IndexArtificial Analysis–56.1
Safety & honesty
MASKScale AI SEAL89.7%†–
Metadata
AA Openness IndexArtificial Analysis–52.8

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

  • MiMo-V2.6-Pro: 1 score revision.
  • GPT-5.4 (xhigh): 1 new result (GDPval-AA v2.1).
  • MiMo-V2.6-Pro: 1 score revision.
  • GPT-5.4 (xhigh): 1 new result (MirrorCode).
  • MiMo-V2.6-Pro: 6 score revisions.
  • MiMo-V2.6-Pro: 1 score revision.
  • GPT-5.4 (xhigh): BenchLeader Index 63.7 → 63.2.
  • GPT-5.4 (xhigh): 1 score revision.
  • MiMo-V2.6-Pro: 1 score revision.
  • MiMo-V2.6-Pro: BenchLeader Index 61.1 → 57.2.

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

GPT-5.4 vs MiMo-V2.6-Pro: questions

Is GPT-5.4 better than MiMo-V2.6-Pro?
GPT-5.4 (xhigh) leads on quality: 63.1 vs 57.5. The BenchLeader Index combines every independent quality benchmark; GPT-5.4 (xhigh) is ahead overall as of 16 Nov 2026, but check the category scores for your use.
GPT-5.4 vs MiMo-V2.6-Pro: which is cheaper, and will that change?
MiMo-V2.6-Pro is cheaper today: $0.544 against $5.63 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is GPT-5.4 or MiMo-V2.6-Pro better for coding?
GPT-5.4 scores higher in coding (65 vs 59 on the category index, where 50 is average). On SciCode, the top shared coding benchmark, it is 56.6% for GPT-5.4 against 60.9% for MiMo-V2.6-Pro.
Which is faster, GPT-5.4 or MiMo-V2.6-Pro?
GPT-5.4 streams faster (73 vs 54 output tokens per second, Artificial Analysis 7-day median).
Is GPT-5.4 or MiMo-V2.6-Pro better for agents and tool use?
GPT-5.4 scores higher in agentic tasks (63 vs 57 on the category index, where 50 is average).
More questions (4)
Which is better for long documents?
MiMo-V2.6-Pro scores higher on long-context benchmarks (68 vs 66). Both take about 1.1M tokens: GPT-5.4 1,050,000, MiMo-V2.6-Pro 1,048,576.
Which has lower latency?
MiMo-V2.6-Pro answers first: 42.3 s against 144.9 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 GPT-5.4 vs MiMo-V2.6-Pro?
Across the 13 benchmarks both have, GPT-5.4 leads 0, MiMo-V2.6-Pro leads 13. Independent sources split: GPT-5.4 leads on BenchLeader Index; MiMo-V2.6-Pro leads on Arena Text.
When was MiMo-V2.6-Pro released?
MiMo-V2.6-Pro was released on 21 Sep 2026.