BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

GPT-6 Sol vs MiMo-V2.6-Flash

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

Updated · Data from 10 sources

As of 6 Nov 2026, GPT-6 Sol is ahead of MiMo-V2.6-Flash on quality (63.3 vs 55.0 on the BenchLeader Index), but MiMo-V2.6-Flash is 23× cheaper at $0.175 vs $4.00 per 1M tokens blended. GPT-6 Sol leads in every category we measure both in. GPT-6 Sol streams 3×+ faster.

Key facts (as of 6 Nov 2026)

  • GPT-6 Sol scores 63.3 on the BenchLeader Index and MiMo-V2.6-Flash scores 55.0 (BenchLeader, 6 Nov 2026).
  • GPT-6 Sol costs $4.00 per 1M tokens blended (list price, 6 Nov 2026).
  • MiMo-V2.6-Flash costs $0.175 per 1M tokens blended (list price, 6 Nov 2026).
  • GPT-6 Sol streams 3×+ faster than MiMo-V2.6-Flash (Artificial Analysis, 7-day median, 6 Nov 2026).
  • Of the 49 benchmarks both models have, GPT-6 Sol leads 38 (BenchLeader, 6 Nov 2026).
  • On Arena WebDev, GPT-6 Sol is rated 1678 and MiMo-V2.6-Flash 1637 (LMArena, as published by 6 Nov 2026).
GPT-6 Sol benchmarks max OpenAI MiMo-V2.6-Flash 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
CodingGPT-6 Sol67 vs 57 category score; Arena WebDev 1678 vs 1637High
Agents & tool useGPT-6 Sol59 vs 57 category score; Arena Agent 6.8 vs 0.2Medium
MathsGPT-6 Sol69 vs 63 category score; Arena Maths 1448 vs 1463Medium
ReasoningGPT-6 Sol67 vs 48 category score; HLE-Diamond (AA) 45.2% vs 13.7%High
KnowledgeGPT-6 Sol62 vs 39 category score; AA-Omniscience v1.1 21.4 vs -10.8High
Long documentsGPT-6 Sol67 vs 62 category score; AA-LCR v1.1 83.7% vs 74.3%Medium
Images and other mediaGPT-6 Sol61 vs 57 category score; MMMU-Pro 83.0% vs 73.1%Medium
Chat quality (human preference)GPT-6 Sol64 vs 63 on Arena votesMedium
Lowest costMiMo-V2.6-Flash$0.175 vs $4.00 per 1M blended; agentic coding session $0.009 vs $0.160–
Fastest first answerEither66.6 s (GPT-6 Sol (max)) vs 53.7 s (MiMo-V2.6-Flash) 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-6 Sol (max) leads on quality: 63.3 vs 55.0.
  • GPT-6 Sol (max) is stronger in agents & tools, coding, human preference, knowledge, long context, maths, multimodal, reasoning.
  • MiMo-V2.6-Flash is 23× cheaper ($0.175 vs $4.00 per 1M blended).
  • GPT-6 Sol (max) streams 3.2× faster (111 vs 35 tokens per second).

At a glance

SpecGPT-6 SolMiMo-V2.6-Flash
DeveloperOpenAIXiaomi
Released22 Sep 202621 Sep 2026
Context window (tokens)1,050,0001,048,576
Compared atGPT-6 Sol (max)MiMo-V2.6-Flash
Open weightsNoYes
Price in / out / cached input per 1M$2.00 / $10.00 / $0.200$0.140 / $0.280 / $0.0028

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-6 Sol (max)MiMo-V2.6-FlashLeader
Reasoning67.248.3GPT-6 Sol
Coding67.157.4GPT-6 Sol
Agents & tools59.457.2GPT-6 Sol
Maths68.862.8GPT-6 Sol
Knowledge62.339.3GPT-6 Sol
Human preference64.463.1GPT-6 Sol
Multimodal61.257.1GPT-6 Sol
Long context66.761.9GPT-6 Sol

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 contextMultimodalHumanpreferenceGPT-6 Sol, Reasoning: 67.2GPT-6 Sol, Coding: 67.1GPT-6 Sol, Agents & tools: 59.4GPT-6 Sol, Maths: 68.8GPT-6 Sol, Knowledge: 62.3GPT-6 Sol, Long context: 66.7GPT-6 Sol, Multimodal: 61.2GPT-6 Sol, Human preference: 64.4MiMo-V2.6-Flash, Reasoning: 48.3MiMo-V2.6-Flash, Coding: 57.4MiMo-V2.6-Flash, Agents & tools: 57.2MiMo-V2.6-Flash, Maths: 62.8MiMo-V2.6-Flash, Knowledge: 39.3MiMo-V2.6-Flash, Long context: 61.9MiMo-V2.6-Flash, Multimodal: 57.1MiMo-V2.6-Flash, Human preference: 63.1

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

ModelIndex

Where sources disagree

Across the 49 benchmarks both have, GPT-6 Sol leads 38, MiMo-V2.6-Flash leads 11. Every overall index that rates both puts GPT-6 Sol ahead.

Overall indexGPT-6 Sol (max)MiMo-V2.6-FlashLeader
BenchLeader Index63.355.0GPT-6 Sol
AA Intelligence Index v5.044.226.7GPT-6 Sol
Arena Text14581448GPT-6 Sol

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-6 Sol (max)MiMo-V2.6-Flash
Blended price per 1M tokens$4.00$0.175
Input / output per 1M$2.00 / $10.00$0.140 / $0.280
Cached input per 1M$0.200$0.0028
Price per task: chat reply$0.0038$0.0001
Price per task: summarize a 30-page report$0.030$0.0018
Price per task: code edit$0.027$0.0013
Price per task: agentic coding session$0.160$0.0095
Price per task: structured extraction$0.0060$0.0003
Agentic coding session, input mostly cached$0.079$0.0033
Cost per benchmark task (AA Index)$2.47$0.129
Output tokens per task (AA Index)54k132k
Of which reasoning35k100k
Time per task (AA Index)8 min63 min
Output speed111 tok/s35 tok/s
Time to first answer66.6 s53.7 s
Time to a complete answer143.4 s257.8 s

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $2.47 on GPT-6 Sol (max) and $0.129 on MiMo-V2.6-Flash, and MiMo-V2.6-Flash wrote 2.4× as many output tokens for the same work, which a per-token price hides.

Each model is compared at its best-ranked reasoning-effort setting. GPT-6 Sol (max) scores 63.3 and takes 67 s to its first answer token; GPT-6 Sol (xhigh) scores 58.7 in 23 s, GPT-6 Sol (high) scores 58.3 in 11 s. MiMo-V2.6-Flash scores 55.0 and takes 54 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

108 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-6 Sol (max)MiMo-V2.6-Flash
BenchLeader IndexBenchLeader63.355.0
Agents & tools scoreBenchLeader59.457.2
Coding scoreBenchLeader67.157.4
Human preference scoreBenchLeader64.463.1
Knowledge scoreBenchLeader62.339.3
Long context scoreBenchLeader66.761.9
Maths scoreBenchLeader68.862.8
Multimodal scoreBenchLeader61.257.1
Reasoning scoreBenchLeader67.248.3
Blended price $/MBenchLeader$4.00$0.175
Output speedBenchLeader111 tok/s35 tok/s
Time to first answerBenchLeader66.6 s53.7 s
Context windowBenchLeader1,050,000 tokens1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub94.3%–
Arena Hard PromptsArena (LMArena)14841475
LiveBench 1.0 ReasoningLiveBench88.7%–
Humanity's Last Exam (AA)Artificial Analysis47.9%35.1%
HLE-Diamond (AA)Artificial Analysis45.2%13.7%
ARC-AGI-1ARC Prize95.5%–
ARC-AGI-2ARC Prize89.6%–
ARC-AGI-3ARC Prize23.0%–
CritPtArtificial Analysis30.9%12.0%
CritPt v2 hard subset (AA)Artificial Analysis66.5%29.5%
MysteryMechanismVals AI30.2%21.6%
Chess PuzzlesEpoch AI Benchmarking Hub56.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub56.0%–
LMCAEpoch AI Benchmarking Hub59.1%–
DTBenchEpoch AI Benchmarking Hub97.3%–
Coding
SciCodeSciCode57.6%51.3%
FrontierCodeCognition49.3%–
Arena CodingArena (LMArena)15191510
Arena WebDevArena (LMArena)16781637
LiveBench 1.0 CodingLiveBench81.8%–
IOIVals AI82.6%47.7%
Code MigrationVals AI57.2%40.9%
Code Migration v1.1Vals AI–39.5%
ProgramBenchVals AI2.0%0.5%
Vibe Code Bench v1.1Vals AI87.8%79.0%
Show all 108 benchmarks
MetricPublisherGPT-6 Sol (max)MiMo-V2.6-Flash
Coding
ALE-BenchEpoch AI Benchmarking Hub24621244.2
Terminal-Bench 4.0 (AA)Artificial Analysis43.9%22.7%
GameBench 2 OverallDuelLab GameBench 272.9†–
GameBench 2 by reasoning settingDuelLab GameBench 279.3†39.4†
FrontierCode 1.1 (AA)Artificial Analysis–42.1%
FrontierCode 2.0 (AA)Artificial Analysis35.6%21.6%
Agents & tools
Terminal-BenchTerminal-Bench49.4%–
APEX-AgentsMercor54.3%–
Arena AgentArena (LMArena)6.80.2
LiveBench Agentic CodingLiveBench52.9%–
Terminal-Bench 2.1 (Vals)Vals AI83.2%76.4%
GDPval-AA v2.1, normalised scoreArtificial Analysis–55.4%
GDPval-AA v2.1Artificial Analysis–1614
ITBench SRE (AA)Artificial Analysis49.4%–
CyberBenchVals AI78.0%75.4%
Finance Agent v2Vals AI49.0%56.3%
Finance Agent v2.1Vals AI–63.3%
Harvey's Legal Agent Benchmark v1.1Vals AI–13.3%
Harvey's Legal Agent Benchmark v1Vals AI1.7%11.3%
Legal Research BenchVals AI28.9%38.0%
SREBenchVals AI30.5%4.2%
Tax Agent BenchVals AI15.0%28.5%
Terminal-Bench 4.0 (Vals)Vals AI44.4%24.2%
Terminal-Bench ScienceVals AI30.0%4.3%
Arena SearchArena (LMArena)1228–
Vending-Bench 2Andon Labs14427.9†–
AutomationBenchArtificial Analysis61.6%64.0%
GDP.pdf (AA)Artificial Analysis25.2%9.0%
Harvey LAB-AA v1.1Harvey3.6%–
AA-Briefcase v1.1Artificial Analysis14761487
Terminal-Bench-Science 0.1 (AA)Artificial Analysis30.0%4.3%
CWE-Bench (AA)Artificial Analysis64.2%–
DeepSec-Bench (AA)Artificial Analysis46.3%–
CyberGym end-to-end (AA)Artificial Analysis0.0%–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub89.8%–
FrontierMath Tier 4Epoch AI Benchmarking Hub90.0%–
OTIS Mock AIMEEpoch AI Benchmarking Hub100.0%–
ProofBenchVals AI83.0%†63.0%
LiveBench MathematicsLiveBench96.4%–
Arena MathsArena (LMArena)14481463
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub60.7%–
LiveBench 1.0 Data AnalysisLiveBench81.2%–
AA-Omniscience v1.1Artificial Analysis21.4-10.8
PRBench LegalScale AI SEAL38.6%†–
BioMysteryBenchVals AI74.8%69.3%
Excel Modeling BenchmarkVals AI71.5%65.5%
Excel Modeling Benchmark v1.1Vals AI–68.9%
MedCodeVals AI47.1%41.1%
MedScribeVals AI82.0%85.3%
Public Benefits Bench v1.1Vals AI56.6%67.6%
EBR-benchEpoch AI Benchmarking Hub53.3%–
GDP.pdfEpoch AI Benchmarking Hub26.4%–
AA-Omniscience v1.1: accuracyArtificial Analysis51.1%27.2%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis39.1%47.8%
HealthBench Professional (AA)OpenAI62.3%–
Instruction following
LiveBench LanguageLiveBench85.3%–
LiveBench Instruction FollowingLiveBench68.6%–
Arena Instruction FollowingArena (LMArena)14491438
Human preference
Arena TextArena (LMArena)14581448
Arena Creative WritingArena (LMArena)14351393
Arena Multi-turnArena (LMArena)14661445
Arena Longer QueriesArena (LMArena)14671460
Multimodal
Arena VisionArena (LMArena)12401256
MMMU-ProArtificial Analysis83.0%73.1%
SAGEVals AI–83.3%
Blueprint-Bench 2Epoch AI Benchmarking Hub36.9%†–
Code2Video-AAArtificial Analysis–1052
AA-3DArtificial Analysis1121–
Long context
AA-LCR v1.1Artificial Analysis83.7%74.3%
MLCRArtificial Analysis33.9%11.7%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub164.1†–
LiveBench 1.0LiveBench79.3%–
AA Intelligence Index v5.0Artificial Analysis44.226.7
AA Intelligence Index v4.3.2Artificial Analysis47.637.9
AA Progress Index v1Artificial Analysis334311.8
Vals IndexVals AI–54.8
Vals Finance IndexVals AI–57.1%
Vals RSI IndexVals AI28.1%–
Vals Science IndexVals AI45.0%31.7%
Vals Router IndexVals AI–54.8%
AA Cyber IndexArtificial Analysis36.8–
Safety & honesty
FORTRESSScale AI SEAL13.5%†–
Metadata
AA Openness IndexArtificial Analysis–52.8

Data as of 2026-11-06. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261103-09b2218cbd85 (https://benchmarks.duellab.org/releases/gb2-20261103-09b2218cbd85/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • GPT-6 Sol (max): 1 score revision.
  • GPT-6 Sol (max): 6 new results (AA Intelligence Index v5.0, AA-Omniscience v1.1, CritPt v2 hard subset (AA)…).
  • GPT-6 Sol (max): 1 score revision.
  • MiMo-V2.6-Flash: 1 score revision.
  • GPT-6 Sol (max): 1 score revision.
  • MiMo-V2.6-Flash: 1 new result (AA-Briefcase v1.1); 1 score revision.
  • GPT-6 Sol (max): BenchLeader Index 64.1 → 63.5.
  • MiMo-V2.6-Flash: BenchLeader Index 58.7 → 55.0.
  • GPT-6 Sol (max): 2 new results (AA Progress Index v1, HLE-Diamond (AA)).
  • MiMo-V2.6-Flash: 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; MiMo-V2.6-Flash cost per task moved from $0.062/M to $0.13/M (106% up).

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

GPT-6 Sol vs MiMo-V2.6-Flash: questions

Is GPT-6 Sol better than MiMo-V2.6-Flash?
GPT-6 Sol (max) leads on quality: 63.3 vs 55.0. The BenchLeader Index combines every independent quality benchmark; GPT-6 Sol (max) is ahead overall as of 6 Nov 2026, but check the category scores for your use.
GPT-6 Sol vs MiMo-V2.6-Flash: which is cheaper, and will that change?
MiMo-V2.6-Flash is cheaper today: $0.175 against $4.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is GPT-6 Sol or MiMo-V2.6-Flash better for coding?
GPT-6 Sol scores higher in coding (67 vs 57 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1678 for GPT-6 Sol against 1637 for MiMo-V2.6-Flash.
Which is faster, GPT-6 Sol or MiMo-V2.6-Flash?
GPT-6 Sol streams 3×+ faster (111 vs 35 output tokens per second, Artificial Analysis 7-day median).
Is GPT-6 Sol or MiMo-V2.6-Flash better for agents and tool use?
GPT-6 Sol scores higher in agentic tasks (59 vs 57 on the category index, where 50 is average). On Arena Agent, the top shared agents & tools benchmark, it is 6.8 for GPT-6 Sol against 0.2 for MiMo-V2.6-Flash.
More questions (4)
Which is better for long documents?
GPT-6 Sol scores higher on long-context benchmarks (67 vs 62). Both take about 1.1M tokens: GPT-6 Sol 1,050,000, MiMo-V2.6-Flash 1,048,576.
Which has lower latency?
MiMo-V2.6-Flash answers first: 53.7 s against 66.6 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-6 Sol vs MiMo-V2.6-Flash?
Across the 49 benchmarks both have, GPT-6 Sol leads 38, MiMo-V2.6-Flash leads 11. Every overall index that rates both puts GPT-6 Sol ahead.
When was GPT-6 Sol released?
GPT-6 Sol was released on 22 Sep 2026.