BenchLeader

GPT-5.5 Pro vs MiMo-V2.5-Pro

Full profiles: GPT-5.5 Pro (OpenAI) and MiMo-V2.5-Pro (Xiaomi): every benchmark, price and rank.

Updated · Data from 9 sources

As of 16 Nov 2026, there are not enough shared benchmarks for a verdict, and MiMo-V2.5-Pro is 45× cheaper at $1.50 vs $67.50 per 1M tokens blended. GPT-5.5 Pro leads in every category we measure both in. MiMo-V2.5-Pro streams 3×+ faster.

Key facts (as of 16 Nov 2026)

  • GPT-5.5 Pro costs $67.50 per 1M tokens blended (list price, 16 Nov 2026).
  • MiMo-V2.5-Pro costs $1.50 per 1M tokens blended (list price, 16 Nov 2026).
  • MiMo-V2.5-Pro streams 3×+ faster than GPT-5.5 Pro (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 2 benchmarks both models have, GPT-5.5 Pro leads 2 (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
MathsGPT-5.5 Pro70 vs 58 category scoreHigh
ReasoningGPT-5.5 Pro68 vs 66 category score; LMCA 53.9% vs 29.5%Medium
Chat quality (human preference)–Only MiMo-V2.5-Pro has an Arena rating (60 category score)Not comparable
Lowest costMiMo-V2.5-Pro$1.50 vs $67.50 per 1M blended; agentic coding session $0.072 vs $2.52–
Fastest first answerMiMo-V2.5-Pro2.8 s (MiMo-V2.5-Pro) vs 6.0 s (GPT-5.5 Pro (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

  • Not enough shared benchmarks for a verdict.
  • GPT-5.5 Pro (xhigh) is stronger in maths, reasoning.
  • MiMo-V2.5-Pro is 45× cheaper ($1.50 vs $67.50 per 1M blended).
  • MiMo-V2.5-Pro streams 3.2× faster (68 vs 21 tokens per second).

At a glance

SpecGPT-5.5 ProMiMo-V2.5-Pro
DeveloperOpenAIXiaomi
Released23 Apr 202623 Apr 2026
Context window (tokens)1,050,0001,048,576
Compared atGPT-5.5 Pro (xhigh)MiMo-V2.5-Pro
Open weightsNoYes
Price in / out / cached input per 1M$30.00 / $180.00 / –$1.00 / $3.00 / $0.0043

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-5.5 Pro (xhigh)MiMo-V2.5-ProLeader
Reasoning67.965.8GPT-5.5 Pro
Coding–56.2not comparable
Agents & tools–46.0not comparable
Maths70.057.6GPT-5.5 Pro
Knowledge–54.5not comparable
Human preference–60.1not 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 & toolsMathsKnowledgeHumanpreferenceGPT-5.5 Pro, Reasoning: 67.9GPT-5.5 Pro, Maths: 70.0MiMo-V2.5-Pro, Reasoning: 65.8MiMo-V2.5-Pro, Coding: 56.2MiMo-V2.5-Pro, Agents & tools: 46.0MiMo-V2.5-Pro, Maths: 57.6MiMo-V2.5-Pro, Knowledge: 54.5MiMo-V2.5-Pro, Human preference: 60.1

Dashed: that model has no result on the axis it crosses; the line only joins its neighbours.

Not drawn, no result for these models: Instruction following, Long context, Multimodal.

Model, click to add or removeIndex

Where sources disagree

Across the 2 benchmarks both have, GPT-5.5 Pro leads 2, MiMo-V2.5-Pro leads 0.

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-5.5 Pro (xhigh)MiMo-V2.5-Pro
Blended price per 1M tokens$67.50$1.50
Input / output per 1M$30.00 / $180.00$1.00 / $3.00
Cached input per 1M–$0.0043
Price per task: chat reply$0.066$0.0013
Price per task: summarize a 30-page report$0.468$0.014
Price per task: code edit$0.450$0.011
Price per task: agentic coding session$2.52$0.072
Price per task: structured extraction$0.096$0.0026
Agentic coding session, input mostly cached–$0.027
Output speed21 tok/s68 tok/s
Time to first answer6.0 s2.8 s

Each model is compared at its best-ranked reasoning-effort setting. GPT-5.5 Pro (xhigh) and takes 6 s to its first answer token. MiMo-V2.5-Pro scores 55.5 and takes 3 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

66 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-5.5 Pro (xhigh)MiMo-V2.5-Pro
BenchLeader IndexBenchLeader–55.5
Maths scoreBenchLeader70.057.6
Reasoning scoreBenchLeader67.965.8
Agents & tools scoreBenchLeader–46.0
Coding scoreBenchLeader–56.2
Human preference scoreBenchLeader–60.1
Knowledge scoreBenchLeader–54.5
Blended price $/MBenchLeader$67.50$1.50
Output speedBenchLeader21 tok/s68 tok/s
Time to first answerBenchLeader6.0 s2.8 s
Context windowBenchLeader1,050,000 tokens1,048,576 tokens
Reasoning
SimpleBenchSimpleBench76.9%†–
Arena Hard PromptsArena (LMArena)–1494
GPQA Diamond (AA)Artificial Analysis–76.2%†
Humanity's Last Exam (AA)Artificial Analysis–14.8%†
GPQA Diamond (Vals)Vals AI–82.6%
ARC-AGI-1ARC Prize95.0%–
ARC-AGI-2ARC Prize84.2%–
CritPtArtificial Analysis30.6%1.1%†
Arena ExpertArena (LMArena)–1507
Chess PuzzlesEpoch AI Benchmarking Hub62.0%–
LMCAEpoch AI Benchmarking Hub53.9%29.5%
DTBenchEpoch AI Benchmarking Hub96.0%84.5%
Coding
SciCodeSciCode–50.2%
Arena CodingArena (LMArena)–1520
Arena WebDevArena (LMArena)–1479
SciCode (AA)Artificial Analysis–50.6%†
LiveCodeBenchVals AI–81.3%
SWE-bench (Vals)Vals AI–74.0%
Code MigrationVals AI–21.6%
Vibe Code Bench v1.1Vals AI–34.1%
ALE-BenchEpoch AI Benchmarking Hub–899.8
Terminal-Bench 4.0 (AA)Artificial Analysis–0.0%†
Terminal-Bench 2.1 (AA)Artificial Analysis–65.2%†
GameBench 2 by reasoning settingDuelLab GameBench 2–27.4†
Design Arena WebsiteDesign Arena–1262
Show all 66 benchmarks
MetricPublisherGPT-5.5 Pro (xhigh)MiMo-V2.5-Pro
Agents & tools
Terminal-Bench HardArtificial Analysis–35.6%†
τ²-Bench Telecom (AA)Artificial Analysis–72.5%†
Terminal-Bench 2.1 (Vals)Vals AI–57.3%
GDPval-AA v2.1Artificial Analysis–1128†
τ³-Banking (AA)Artificial Analysis–9.9%†
ITBench SRE (AA)Artificial Analysis–38.2%†
Analyst Agent (AA)Artificial Analysis–20.0%†
APEX-Agents (AA)Artificial Analysis–2.4%†
Finance Agent v2Vals AI–41.5%
Harvey's Legal Agent Benchmark v1Vals AI–2.1%
Legal Research BenchVals AI–15.9%
Tax Agent BenchVals AI–11.5%
Terminal-Bench 4.0 (Vals)Vals AI–0.5%
AutomationBenchArtificial Analysis–13.7%†
GDP.pdf (AA)Artificial Analysis–4.0%†
AA-Briefcase v1.1Artificial Analysis–884†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub87.7%–
FrontierMath Tier 4Epoch AI Benchmarking Hub78.0%–
ProofBenchVals AI–22.0%
Arena MathsArena (LMArena)–1479
Knowledge
MMLU-ProVals AI–84.6%
LegalBenchVals AI–77.1%
CorpFinVals AI–61.4%
TaxEvalVals AI–73.8%
Excel Modeling BenchmarkVals AI–55.2%
MedCodeVals AI–32.5%
MedScribeVals AI–83.7%
Instruction following
IFBenchArtificial Analysis–42.7%†
Arena Instruction FollowingArena (LMArena)–1468
Human preference
Arena TextArena (LMArena)–1466
EQ-Bench 4EQ-Bench–1198
EQ-Bench Creative Writing v3EQ-Bench–1493
Arena Creative WritingArena (LMArena)–1433
Arena Multi-turnArena (LMArena)–1480
Arena Longer QueriesArena (LMArena)–1486
Long context
AA-LCR v1.1Artificial Analysis–41.7%†
MLCRArtificial Analysis–5.6%†
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub162.7†–
AA Intelligence Index v4.3.2Artificial Analysis–18.3†
AA Progress Index v1Artificial Analysis–241†
Metadata
AA Openness IndexArtificial Analysis–38.9†

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-V2.5-Pro: MiMo-V2.5-Pro cached input price moved from $0.20/M to $0.0043/M (98% down).
  • MiMo-V2.5-Pro: BenchLeader Index 55.0 → 55.6.
  • MiMo-V2.5-Pro: 2 score revisions.
  • MiMo-V2.5-Pro: 1 new result (Harvey's Legal Agent Benchmark v1).

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

GPT-5.5 Pro vs MiMo-V2.5-Pro: questions

GPT-5.5 Pro vs MiMo-V2.5-Pro: which is cheaper, and will that change?
MiMo-V2.5-Pro is cheaper today: $1.50 against $67.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Which is faster, GPT-5.5 Pro or MiMo-V2.5-Pro?
MiMo-V2.5-Pro streams 3×+ faster (68 vs 21 output tokens per second, Artificial Analysis 7-day median).
Which is better for long documents?
Both take about 1.1M tokens: GPT-5.5 Pro 1,050,000, MiMo-V2.5-Pro 1,048,576.
Which has lower latency?
MiMo-V2.5-Pro answers first: 2.8 s against 6.0 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-5.5 Pro vs MiMo-V2.5-Pro?
Across the 2 benchmarks both have, GPT-5.5 Pro leads 2, MiMo-V2.5-Pro leads 0.
More questions (1)
When was GPT-5.5 Pro released?
GPT-5.5 Pro was released on 23 Apr 2026.