BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases9

GPT-5.5 vs Grok 4.6

Full profiles: GPT-5.5 (OpenAI) and Grok 4.6 (SpaceXAI): every benchmark, price and rank.

Updated · Data from 11 sources

As of 25 Oct 2026, GPT-5.5 (xhigh) and Grok 4.6 (medium) are neck and neck on quality (66.1 vs 64.2 on the BenchLeader Index, inside the ±5.7 margin of error), and Grok 4.6 (medium) costs 3.8× less, which makes it the better buy for most workloads. Pick GPT-5.5 for reasoning (76 vs 64); pick Grok 4.6 for agents & tools (74 vs 62).

GPT-5.5 benchmarks xhigh OpenAI Grok 4.6 benchmarks medium SpaceXAIOpen in the comparison tool

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
CodingGrok 4.662 vs 61 category score; SciCode 55.9% vs 55.8%Medium
Agents & tool useGrok 4.674 vs 62 category score; τ³-Banking (AA) 44.3% vs 39.0%High
ReasoningGPT-5.576 vs 64 category score; ARC-AGI-2 85.0% vs 61.3%High
KnowledgeGrok 4.674 vs 65 category score; AA-Omniscience 28 vs 20.5High
Long documentsGPT-5.567 vs 66 category score; AA-LCR 84.3% vs 81.0%Medium
Lowest costGrok 4.6$3.00 vs $11.25 per 1M blended; agentic coding session $0.144 vs $0.420–
Fastest first answerGrok 4.625.5 s (Grok 4.6 (medium)) vs 39.1 s (GPT-5.5 (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-5.5 (xhigh) and Grok 4.6 (medium) are neck and neck on quality (66.1 vs 64.2, within the ±5.7 margin).
  • GPT-5.5 (xhigh) is stronger in instruction following, long context, maths, multimodal, reasoning.
  • Grok 4.6 (medium) is stronger in agents & tools, coding, composite, knowledge.
  • Grok 4.6 (medium) is 3.8× cheaper ($3.00 vs $11.25 per 1M blended).

At a glance

SpecGPT-5.5Grok 4.6
DeveloperOpenAISpaceXAI
Released23 Apr 202611 Aug 2026
Context window (tokens)1,050,000500,000
Compared atGPT-5.5 (xhigh)Grok 4.6 (medium)
Open weightsNoNo
Price in / out / cached input per 1M$5.00 / $30.00 / $0.500$2.00 / $6.00 / $0.500

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 (xhigh)Grok 4.6 (medium)Leader
Reasoning75.564.4GPT-5.5
Coding61.162.3Grok 4.6
Agents & tools62.373.5Grok 4.6
Maths69.7–not comparable
Knowledge65.374.1Grok 4.6
Instruction following73.9–not comparable
Multimodal62.9–not comparable
Long context67.365.6GPT-5.5
Composite72.477.4Grok 4.6

Where sources disagree

Across the 17 benchmarks both have, GPT-5.5 leads 9, Grok 4.6 leads 6 and 2 are tied. Independent sources split: GPT-5.5 leads on Arena Text; Grok 4.6 leads on AA Intelligence Index v4.3.2.

Overall indexGPT-5.5 (xhigh)Grok 4.6 (medium)Leader
BenchLeader Index66.164.2neck and neck
AA Intelligence Index v4.3.238.442.8Grok 4.6
Arena Text1482†1456GPT-5.5

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; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.

MeasureGPT-5.5 (xhigh)Grok 4.6 (medium)
Blended price per 1M tokens$11.25$3.00
Input / output per 1M$5.00 / $30.00$2.00 / $6.00
Cached input per 1M$0.500$0.500
Price per task: chat reply$0.011$0.0026
Price per task: summarize a 30-page report$0.078$0.028
Price per task: code edit$0.075$0.021
Price per task: agentic coding session$0.420$0.144
Price per task: structured extraction$0.016$0.0052
Cost to run a full benchmark suite$2.63$1.14
Output speed86 tok/s78 tok/s
Time to first answer39.1 s25.5 s
Time to a complete answer45.0 s31.9 s

Per token is not per task: on Artificial Analysis' full suite GPT-5.5 (xhigh) cost $2.63 and Grok 4.6 (medium) $1.14, because Grok 4.6 (medium) used about 1.6× as many tokens for the same work, which a per-token price hides.

Each model is compared at its best-ranked reasoning-effort setting. GPT-5.5 (xhigh) scores 66.1 and takes 39 s to its first answer token; GPT-5.5 (high) scores 65.3 in 14 s, GPT-5.5 (medium) scores 62.9 in 3 s. Grok 4.6 (medium) scores 64.2 and takes 25 s to its first answer token; Grok 4.6 (high) scores 62.9 in 25 s, Grok 4.6 (xhigh) scores 62.5 in 29 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

144 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 (xhigh)Grok 4.6 (medium)
BenchLeader IndexBenchLeader66.164.2
Agents & tools scoreBenchLeader62.373.5
Coding scoreBenchLeader61.162.3
Composite scoreBenchLeader72.477.4
Instruction following scoreBenchLeader73.9–
Knowledge scoreBenchLeader65.374.1
Long context scoreBenchLeader67.365.6
Maths scoreBenchLeader69.7–
Multimodal scoreBenchLeader62.9–
Reasoning scoreBenchLeader75.564.4
Blended price $/MBenchLeader$11.25$3.00
Output speedBenchLeader86 tok/s78 tok/s
Time to first answerBenchLeader39.1 s25.5 s
Context windowBenchLeader1,050,000 tokens500,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub90.7%†94.0%†
SimpleBenchSimpleBench69.0%†75.9%†
Arena Hard PromptsArena (LMArena)1499†1482†
LiveBench 2.0 ReasoningLiveBench89.7%†91.2%†
LiveBench 1.0 ReasoningLiveBench89.7%90.5%†
GPQA Diamond (AA)Artificial Analysis93.5%93.5%
Humanity's Last Exam (AA)Artificial Analysis45.8%42.1%
GPQA Diamond (Vals)Vals AI93.2%94.7%†
Kagi LLM BenchmarkKagi LLM Benchmark88.8%†–
ARC-AGI-1ARC Prize95.0%87.5%
ARC-AGI-2ARC Prize85.0%61.3%
ARC-AGI-3ARC Prize0.4%†2.1%†
CritPtArtificial Analysis27.1%17.7%
MysteryMechanismVals AI–30.6%†
Chess PuzzlesEpoch AI Benchmarking Hub63.0%33.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub56.0%34.0%†
CL-bench LifeEpoch AI Benchmarking Hub22.2%†–
LMCAEpoch AI Benchmarking Hub54.3%48.5%†
DTBenchEpoch AI Benchmarking Hub96.0%97.3%†
ForecastBenchForecastBench64.4%†65.5%†
BTF-3Epoch AI Benchmarking Hub14.3%†–
Coding
SciCodeSciCode55.8%55.9%
WeirdMLWeirdML84.9%67.3%†
FrontierCodeCognition43.0%†48.0%†
GSO-BenchGSO-Bench40.2%–
Show all 144 benchmarks
MetricPublisherGPT-5.5 (xhigh)Grok 4.6 (medium)
Coding
Arena CodingArena (LMArena)1511†1510†
Arena WebDevArena (LMArena)15131617†
LiveBench 2.0 CodingLiveBench85.1%†89.1%†
LiveBench 1.0 CodingLiveBench82.2%76.8%†
LiveCodeBenchVals AI85.3%88.2%†
IOIVals AI–47.6%†
SWE-bench (Vals)Vals AI82.6%95.6%†
Code MigrationVals AI45.2%44.6%†
ProgramBenchVals AI0.5%1.0%†
Vibe Code Bench 1-100Vals AI–14.8%†
Vibe Code Bench v1.1Vals AI69.8%76.2%†
SWE Atlas: Codebase QnAScale AI SEAL45.4%–
SWE Atlas: RefactoringScale AI SEAL44.8%–
SWE Atlas: Test WritingScale AI SEAL42.6%–
MirrorCodeEpoch AI Benchmarking Hub10.0%†–
Surface Evolver BenchEpoch AI Benchmarking Hub88.1%†–
DeepSWE v1.1Epoch AI Benchmarking Hub67.0%67.5%
CursorBenchCursor–36.1%
ALE-BenchEpoch AI Benchmarking Hub1943.01508.1†
FrontierSWEEpoch AI Benchmarking Hub–25.3%†
Terminal-Bench 4.0 (AA)Artificial Analysis14.7%13.1%
Terminal-Bench 2.1 (AA)Artificial Analysis84.3%84.3%
Startrise FrontendStartrise–72.1%†
Agents & tools
Terminal-BenchTerminal-Bench84.7%†20.3%†
OSWorld-Verified 2.0OSWorld13.0%–
Remote Labor IndexScale AI / CAIS6.3%†–
APEX-AgentsMercor55.1%†65.3%†
Arena AgentArena (LMArena)–0.6†
LiveBench Long AgenticLiveBench44.6%†40.7%†
LiveBench BrowserLiveBench48.0%†64.0%†
LiveBench Agentic CodingLiveBench54.0%57.0%†
Terminal-Bench HardArtificial Analysis60.6%–
τ²-Bench Telecom (AA)Artificial Analysis93.9%–
Terminal-Bench 2.1 (Vals)Vals AI76.4%†78.3%†
MCP AtlasScale AI SEAL75.3%–
HiL-BenchScale AI SEAL39.7%†–
τ³-bench Bankingτ²-bench44.6%–
τ³-Banking (AA)Artificial Analysis39.0%44.3%
ITBench SRE (AA)Artificial Analysis45.8%–
Analyst Agent (AA)Artificial Analysis50.0%41.3%†
APEX-Agents (AA)Artificial Analysis37.7%–
CyberBenchVals AI–67.7%†
Finance Agent v2Vals AI51.8%53.7%†
Harvey's Legal Agent Benchmark v1Vals AI3.8%15.8%†
Legal Research BenchVals AI40.4%48.1%†
SkillsBenchVals AI62.2%55.8%†
SREBenchVals AI3.8%–
Tax Agent BenchVals AI21.8%33.1%†
Terminal-Bench 2.0 (Vals)Vals AI73.2%†–
Terminal-Bench 4.0 (Vals)Vals AI–17.2%†
Terminal-Bench ScienceVals AI–11.4%†
Time Horizon Index: KSPVals AI8.3%5.8%†
DeepResearch BenchDeepResearch Bench54.0%†–
PostTrainBenchEpoch AI Benchmarking Hub27.2%–
ExploitBenchEpoch AI Benchmarking Hub47.4%†–
GBAEvalEpoch AI Benchmarking Hub53.2%†–
Vending-Bench 2Andon Labs7523.8†9047.0†
AutomationBenchArtificial Analysis47.3%63.1%
GDP.pdf (AA)Artificial Analysis21.2%17.8%
EnterpriseOps-GymArtificial Analysis46.6%48.3%†
AA-Briefcase v1.1Artificial Analysis11361531†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–6.2%†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub85.3%66.0%†
FrontierMath Tier 4Epoch AI Benchmarking Hub72.5%31.7%†
OTIS Mock AIMEEpoch AI Benchmarking Hub84.4%†97.8%†
ProofBenchVals AI50.0%51.0%†
LiveBench MathematicsLiveBench95.9%92.6%†
AIME 2026MathArena100.0%–
HMMT February 2026MathArena98.5%–
USAMO 2026MathArena98.2%–
MathArena ApexMathArena80.2%–
Arena MathsArena (LMArena)1501†1445†
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub63.0%49.3%†
LiveBench 2.0 Data AnalysisLiveBench81.6%†75.3%†
LiveBench ResearchLiveBench61.6%†84.5%†
LiveBench 1.0 Data AnalysisLiveBench81.6%73.9%†
AA-OmniscienceArtificial Analysis20.528
MMLU-ProVals AI88.1%89.4%†
LegalBenchVals AI86.5%86.3%†
CorpFinVals AI68.4%66.2%†
TaxEvalVals AI75.0%71.1%†
BioMysteryBenchVals AI–72.2%†
CaseLaw v2Vals AI66.2%–
Excel Modeling BenchmarkVals AI64.5%62.7%†
MedCodeVals AI49.1%44.7%†
MedScribeVals AI86.9%86.5%†
Public Benefits Bench v1.1Vals AI60.9%66.8%†
DrugDiscoveryBenchScale AI SEAL51.6%–
EBR-benchEpoch AI Benchmarking Hub34.3%30.5%†
GDP.pdfEpoch AI Benchmarking Hub26.0%16.0%†
AA-Omniscience: accuracyArtificial Analysis58.0%41.9%
AA-Omniscience: non-hallucinationArtificial Analysis11.0%76.0%
Instruction following
LiveBench LanguageLiveBench87.4%83.7%†
LiveBench Instruction FollowingLiveBench70.7%71.9%†
IFBenchArtificial Analysis75.8%–
Arena Instruction FollowingArena (LMArena)1475†1452†
Human preference
Arena TextArena (LMArena)1477†1456†
EQ-Bench 4EQ-Bench1294†–
Arena Creative WritingArena (LMArena)1448†1443†
Arena Multi-turnArena (LMArena)1484†1448†
Arena Longer QueriesArena (LMArena)1487†1475†
Multimodal
Arena VisionArena (LMArena)1297†1263†
LiveBench DesignLiveBench70.3%†79.8%†
MMMU-ProArtificial Analysis79.9%–
MMMU-Pro (Vals)Vals AI88.3%–
MortgageTaxVals AI68.8%64.2%†
SAGEVals AI51.5%28.9%†
Blueprint-Bench 2Epoch AI Benchmarking Hub36.2%†33.2%†
Long context
AA-LCRArtificial Analysis84.3%81.0%
Arena DocumentArena (LMArena)1483†1461†
MLCRArtificial Analysis28.9%18.3%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub159.7†156.6†
LiveBench 2.0LiveBench68.7%†74.9%†
LiveBench 1.0LiveBench80.2%78.0%†
AA Intelligence Index v4.3.2Artificial Analysis38.442.8
Vals RSI IndexVals AI16.5%25.5%†
Vals Science IndexVals AI–38.1%†
Vals Multimodal IndexVals AI68.1%–
Safety & honesty
FORTRESSScale AI SEAL16.3%–

Data as of 2026-10-25. Best configuration of each model; every score links to its source on the model pages.

What changed

  • GPT-5.5 (xhigh): 1 new result (AA-Briefcase v1.1).
  • GPT-5.5 (xhigh): 2 new results (Harvey's Legal Agent Benchmark v1, Vals RSI Index).
  • GPT-5.5 (xhigh): 1 score revision.
  • Grok 4.6 (medium): 1 new result (MLCR).
  • GPT-5.5 (xhigh): 6 new results (USAMO 2026, AutomationBench, GDP.pdf (AA)…).
  • Grok 4.6 (medium): 15 new results (AA Intelligence Index v4.3.2, GPQA Diamond (AA), Humanity's Last Exam (AA)…).
  • GPT-5.5 (xhigh): BenchLeader Index 66.2 → 65.7.
  • GPT-5.5 (xhigh): 1 score revision.
  • GPT-5.5 (xhigh): 1 new result (Chess Puzzles); 1 score revision.
  • Grok 4.6 (medium): 1 score revision.

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

GPT-5.5 vs Grok 4.6: questions

Is GPT-5.5 better than Grok 4.6?
GPT-5.5 (xhigh) and Grok 4.6 (medium) are neck and neck on quality (66.1 vs 64.2, within the ±5.7 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2026-10-25, so check the category scores for your use.
Which is cheaper, GPT-5.5 or Grok 4.6, and will that change?
Grok 4.6 is cheaper today: $3.00 against $11.25 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is GPT-5.5 better than Grok 4.6 for coding?
Grok 4.6 scores higher in coding (62 vs 61 on the category index, where 50 is average). On SciCode, the top shared coding benchmark, it is 55.8% for GPT-5.5 against 55.9% for Grok 4.6.
Is GPT-5.5 better than Grok 4.6 for agents and tool use?
Grok 4.6 scores higher in agentic tasks (74 vs 62 on the category index, where 50 is average). On τ³-Banking (AA), the top shared agents & tools benchmark, it is 39.0% for GPT-5.5 against 44.3% for Grok 4.6.
Which is better for long documents?
GPT-5.5 scores higher on long-context benchmarks (67 vs 66). GPT-5.5 accepts more context: 1,050,000 against 500,000 tokens for Grok 4.6.
Which has lower latency?
Grok 4.6 (medium) answers first: 25.5 s against 39.1 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.
Which is faster, GPT-5.5 or Grok 4.6?
GPT-5.5 streams faster: 86 against 78 output tokens per second.
Do independent benchmarks agree on GPT-5.5 vs Grok 4.6?
Across the 17 benchmarks both have, GPT-5.5 leads 9, Grok 4.6 leads 6 and 2 are tied. Independent sources split: GPT-5.5 leads on Arena Text; Grok 4.6 leads on AA Intelligence Index v4.3.2.
When was Grok 4.6 released?
Grok 4.6 was released on 11 Aug 2026.