BenchLeader

Grok 4.20 vs Qwen3.8 Max

Full profiles: Grok 4.20 (SpaceXAI) and Qwen3.8 Max (Alibaba): every benchmark, price and rank.

Updated · Data from 14 sources

As of 16 Nov 2026, Grok 4.20 and Qwen3.8 Max are neck and neck on quality (58.3 vs 58.7 on the BenchLeader Index, inside the ±5.4 margin), and Grok 4.20 is 1.9× cheaper at $1.56 vs $3.00 per 1M tokens blended. Choose Grok 4.20 for reasoning (61 vs 52) and fast first answers (12.1 s vs 119.6 s), where it also streams ~1.5× faster; choose Qwen3.8 Max for maths (65 vs 54), coding (62 vs 53), and agents & tools (59 vs 51).

Key facts (as of 16 Nov 2026)

  • Grok 4.20 scores 58.3 on the BenchLeader Index and Qwen3.8 Max scores 58.7 (BenchLeader, 16 Nov 2026).
  • Grok 4.20 costs $1.56 per 1M tokens blended (list price, 16 Nov 2026).
  • Qwen3.8 Max costs $3.00 per 1M tokens blended (list price, 16 Nov 2026).
  • Grok 4.20 streams ~1.5× faster than Qwen3.8 Max (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 40 benchmarks both models have, Qwen3.8 Max leads 40 (BenchLeader, 16 Nov 2026).
  • On Arena WebDev, Grok 4.20 is rated 1376 and Qwen3.8 Max 1676 (LMArena, as published by 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
CodingQwen3.8 Max62 vs 53 category score; Arena WebDev 1676 vs 1376High
Agents & tool useQwen3.8 Max59 vs 51 category score; Terminal-Bench 2.1 (Vals) 67.4% vs 44.2%High
MathsQwen3.8 Max65 vs 54 category score; Arena Maths 1505 vs 1464High
ReasoningGrok 4.2061 vs 52 category score; Arena Hard Prompts 1489 vs 1507High
KnowledgeQwen3.8 Max58 vs 53 category score; MMLU-Pro 88.6% vs 86.3%Medium
Long documentsQwen3.8 Max65 vs 59 category score; AA-LCR v1.1 80.3% vs 69.0%Medium
Images and other mediaQwen3.8 Max66 vs 58 category score; Arena Vision 1312 vs 1263Medium
Chat quality (human preference)Qwen3.8 Max68 vs 66 on Arena votesMedium
Lowest costGrok 4.20$1.56 vs $3.00 per 1M blended; agentic coding session $0.085 vs $0.144–
Fastest first answerGrok 4.2012.1 s (Grok 4.20 (thinking)) vs 119.6 s (Qwen3.8 Max (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

  • Grok 4.20 (thinking) and Qwen3.8 Max (max) are neck and neck on quality (58.3 vs 58.7, within the ±5.4 margin).
  • Grok 4.20 (thinking) is stronger in reasoning.
  • Qwen3.8 Max (max) is stronger in agents & tools, coding, human preference, knowledge, long context, maths, multimodal.
  • Grok 4.20 (thinking) is 1.9× cheaper ($1.56 vs $3.00 per 1M blended).
  • Grok 4.20 (thinking) streams 1.9× faster (111 vs 59 tokens per second).

At a glance

SpecGrok 4.20Qwen3.8 Max
DeveloperSpaceXAIAlibaba
Released17 Feb 20263 Aug 2026
Context window (tokens)2,000,0001,000,000
Compared atGrok 4.20 (thinking)Qwen3.8 Max (max)
Open weightsNoNo
Price in / out / cached input per 1M$1.25 / $2.50 / $0.200$2.00 / $6.00 / $0.250

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.

CategoryGrok 4.20 (thinking)Qwen3.8 Max (max)Leader
Reasoning60.551.5Grok 4.20
Coding53.361.6Qwen3.8 Max
Agents & tools50.558.5Qwen3.8 Max
Maths53.765.2Qwen3.8 Max
Knowledge53.058.0Qwen3.8 Max
Instruction following78.7–not comparable
Human preference66.067.5Qwen3.8 Max
Multimodal58.365.7Qwen3.8 Max
Long context58.964.8Qwen3.8 Max
Composite–45.0not 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 contextMultimodalHumanpreferenceGrok 4.20, Reasoning: 60.5Grok 4.20, Coding: 53.3Grok 4.20, Agents & tools: 50.5Grok 4.20, Maths: 53.7Grok 4.20, Knowledge: 53.0Grok 4.20, Long context: 58.9Grok 4.20, Multimodal: 58.3Grok 4.20, Human preference: 66.0Qwen3.8 Max, Reasoning: 51.5Qwen3.8 Max, Coding: 61.6Qwen3.8 Max, Agents & tools: 58.5Qwen3.8 Max, Maths: 65.2Qwen3.8 Max, Knowledge: 58.0Qwen3.8 Max, Long context: 64.8Qwen3.8 Max, Multimodal: 65.7Qwen3.8 Max, Human preference: 67.5

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

Model, click to add or removeIndex

Where sources disagree

Across the 40 benchmarks both have, Grok 4.20 leads 0, Qwen3.8 Max leads 40. Every overall index that rates both puts Qwen3.8 Max ahead.

Overall indexGrok 4.20 (thinking)Qwen3.8 Max (max)Leader
BenchLeader Index58.358.7neck and neck
Arena Text14721484Qwen3.8 Max

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.

MeasureGrok 4.20 (thinking)Qwen3.8 Max (max)
Blended price per 1M tokens$1.56$3.00
Input / output per 1M$1.25 / $2.50$2.00 / $6.00
Cached input per 1M$0.200$0.250
Price per task: chat reply$0.0013$0.0026
Price per task: summarize a 30-page report$0.017$0.028
Price per task: code edit$0.011$0.021
Price per task: agentic coding session$0.085$0.144
Price per task: structured extraction$0.0030$0.0052
Agentic coding session, input mostly cached$0.038$0.065
Cost per benchmark task (AA Index)–$10.02
Output tokens per task (AA Index)–171k
Of which reasoning–120k
Time per task (AA Index)–48 min
Output speed111 tok/s59 tok/s
Time to first answer12.1 s119.6 s
Time to a complete answer, last measured 13 Nov 202643.8 s294.1 s

Each model is compared at its best-ranked reasoning-effort setting. Grok 4.20 (thinking) scores 58.3 and takes 12 s to its first answer token. Qwen3.8 Max (max) scores 58.7 and takes 120 s to its first answer token; Qwen3.8 Max (xhigh) scores 56.6 in 3 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

157 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.

MetricPublisherGrok 4.20 (thinking)Qwen3.8 Max (max)
BenchLeader IndexBenchLeader58.358.7
Agents & tools scoreBenchLeader50.558.5
Coding scoreBenchLeader53.361.6
Human preference scoreBenchLeader66.067.5
Instruction following scoreBenchLeader78.7–
Knowledge scoreBenchLeader53.058.0
Long context scoreBenchLeader58.964.8
Maths scoreBenchLeader53.765.2
Multimodal scoreBenchLeader58.365.7
Reasoning scoreBenchLeader60.551.5
Composite scoreBenchLeader–45.0
Blended price $/MBenchLeader$1.56$3.00
Output speedBenchLeader111 tok/s59 tok/s
Time to first answerBenchLeader12.1 s119.6 s
Context windowBenchLeader2,000,000 tokens1,000,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub89.3%92.7%†
Arena Hard PromptsArena (LMArena)14891507
LiveBench 2.0 ReasoningLiveBench–88.2%
LiveBench 1.0 ReasoningLiveBench–88.2%
GPQA Diamond (AA)Artificial Analysis91.1%92.8%
Humanity's Last Exam (AA)Artificial Analysis34.5%43.1%
HLE-Diamond (AA)Artificial Analysis–24.0%
GPQA Diamond (Vals)Vals AI88.6%93.7%
Kagi LLM BenchmarkKagi LLM Benchmark75.0%–
ARC-AGI-1ARC Prize89.5%–
ARC-AGI-2ARC Prize65.1%–
ARC-AGI-3ARC Prize0.1%–
CritPtArtificial Analysis6.6%20.0%
CritPt v2 hard subset (AA)Artificial Analysis–29.5%
MysteryMechanismVals AI–23.9%
Arena ExpertArena (LMArena)14771514
Chess PuzzlesEpoch AI Benchmarking Hub29.0%46.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub–38.0%†
CL-benchEpoch AI Benchmarking Hub22.2%†–
CL-bench LifeEpoch AI Benchmarking Hub11.9%†–
LMCAEpoch AI Benchmarking Hub38.7%†46.2%†
DTBenchEpoch AI Benchmarking Hub90.1%†92.0%†
ForecastBenchForecastBench64.1%–
Coding
SciCodeSciCode–53.2%
WeirdMLWeirdML52.3%†–
Show all 157 benchmarks
MetricPublisherGrok 4.20 (thinking)Qwen3.8 Max (max)
Coding
Arena CodingArena (LMArena)15111523
Arena WebDevArena (LMArena)13761676
LiveBench 2.0 CodingLiveBench–60.9%
LiveBench 1.0 CodingLiveBench–72.9%
LiveCodeBenchVals AI84.3%87.8%
IOIVals AI–68.9%
SWE-bench (Vals)Vals AI72.2%85.6%
Code MigrationVals AI0.3%24.0%
ProgramBenchVals AI–0.0%
SWE-bench MultimodalVals AI–42.5%
Vibe Code Bench 1-100Vals AI–12.8%
Vibe Code Bench v1.1Vals AI4.1%64.7%
SWE Atlas v1.1Scale AI SEAL–64.4%
SWE Atlas v1.1: Codebase QnAScale AI SEAL–81.5%
SWE Atlas v1.1: Test WritingScale AI SEAL–58.9%
SWE Atlas v1.1: RefactoringScale AI SEAL–52.9%
MirrorCodeEpoch AI Benchmarking Hub–52.2%†
DeepSWE v1.1Epoch AI Benchmarking Hub–57.5%†
ALE-BenchEpoch AI Benchmarking Hub1150.3†1679.1†
FrontierSWEEpoch AI Benchmarking Hub–17.8%†
Terminal-Bench 4.0 (AA)Artificial Analysis–38.9%
Terminal-Bench 2.1 (AA)Artificial Analysis–88.8%
Startrise FrontendStartrise–71.3%
GameBench 2 OverallDuelLab GameBench 2–54.2†
GameBench 2 by reasoning settingDuelLab GameBench 2–69.6†
Design Arena WebsiteDesign Arena12211269
FrontierCode 1.1 (AA)Artificial Analysis–43.2%
FrontierCode 2.0 (AA)Artificial Analysis–23.4%
Agents & tools
Terminal-BenchTerminal-Bench57.3%†27.0%
APEX-AgentsMercor–63.3%
Arena AgentArena (LMArena)–0.1
LiveBench Long AgenticLiveBench–26.6%
LiveBench BrowserLiveBench–49.4%
LiveBench Agentic CodingLiveBench–64.7%
Terminal-Bench HardArtificial Analysis37.9%–
τ²-Bench Telecom (AA)Artificial Analysis93.0%–
Terminal-Bench 2.1 (Vals)Vals AI44.2%67.4%
HiL-BenchScale AI SEAL20.0%†–
τ³-bench Bankingτ²-bench–55.1%†
GDPval-AA v2.1, normalised scoreArtificial Analysis–58.3%
GDPval-AA v2.1Artificial Analysis–1674
τ³-Banking (AA)Artificial Analysis–51.3%
ITBench SRE (AA)Artificial Analysis–40.3%
Analyst Agent (AA)Artificial Analysis–45.0%
APEX-Agents (AA)Artificial Analysis14.2%42.4%
CyberBenchVals AI–28.6%
Finance Agent v2Vals AI28.5%50.6%
Harvey's Legal Agent Benchmark v1Vals AI0.0%10.4%
Legal Research BenchVals AI13.9%47.6%
SkillsBenchVals AI–42.0%
Tax Agent BenchVals AI7.0%32.1%
Terminal-Bench 2.0 (Vals)Vals AI40.5%–
Terminal-Bench 4.0 (Vals)Vals AI–34.3%
Terminal-Bench ScienceVals AI–12.9%
Arena SearchArena (LMArena)–1227
Arena Agent Task CompletionArena (LMArena)–4.5
Vending-Bench 2Andon Labs4662.9†–
AutomationBenchArtificial Analysis–56.2%
GDP.pdf (AA)Artificial Analysis–22.8%
EnterpriseOps-GymArtificial Analysis–47.6%
AA-Briefcase v1.1Artificial Analysis–1604
GDPxGDPx by SemiAnalysis–3.5%
Design Arena AgenticDesign Arena10931264
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–11.9%
MKA (AA)Artificial Analysis–37.4%
FACTS SearchGoogle–85.0%
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub44.9%74.7%†
FrontierMath Tier 4Epoch AI Benchmarking Hub17.1%46.3%†
OTIS Mock AIMEEpoch AI Benchmarking Hub92.2%100.0%†
ProofBenchVals AI14.0%58.0%
LiveBench MathematicsLiveBench–91.3%
AIME (Vals)Vals AI96.5%–
Arena MathsArena (LMArena)14641505
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub30.2%47.3%†
LiveBench 2.0 Data AnalysisLiveBench–78.4%
LiveBench ResearchLiveBench–72.8%
LiveBench 1.0 Data AnalysisLiveBench–78.4%
AA-Omniscience v1.1Artificial Analysis–10.5
MMLU-ProVals AI86.3%88.6%
LegalBenchVals AI77.7%83.6%
CorpFinVals AI63.7%65.8%
TaxEvalVals AI74.1%75.5%
MedQAVals AI94.5%–
CaseLaw v2Vals AI54.5%–
Excel Modeling BenchmarkVals AI11.7%60.1%
Excel Modeling Benchmark v1.1Vals AI–66.0%
MedCodeVals AI32.2%40.7%
MedScribeVals AI63.4%85.0%
Public Benefits Bench v1.1Vals AI–67.1%
Access to Justice BenchmarkVals AI–76.4%
Financial Coaching BenchVals AI–75.6%
Immigration BenchVals AI–70.1%
GDP.pdfEpoch AI Benchmarking Hub–23.2%†
AA-Omniscience v1.1: accuracyArtificial Analysis–29.5%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–72.9%
Global-MMLU-Lite (AA)Artificial Analysis89.5%–
FACTS ParametricGoogle–30.8%
Instruction following
LiveBench LanguageLiveBench–79.7%
LiveBench Instruction FollowingLiveBench–74.1%
IFBenchArtificial Analysis81.2%–
Arena Instruction FollowingArena (LMArena)14481478
Human preference
Arena TextArena (LMArena)14721484
EQ-Bench Creative Writing v3EQ-Bench1574†–
Arena Creative WritingArena (LMArena)14451472
Arena Multi-turnArena (LMArena)14791495
Arena Longer QueriesArena (LMArena)14661496
Multimodal
Arena VisionArena (LMArena)12631312
LiveBench DesignLiveBench–76.6%
MMMU-ProArtificial Analysis74.6%82.8%
MMMU-Pro (Vals)Vals AI83.5%88.0%
MortgageTaxVals AI45.4%64.0%
SAGEVals AI–81.8%
Humanity's Sixth SenseScale AI SEAL–33.0%
Blueprint-Bench 2Epoch AI Benchmarking Hub0.0%†–
Furniture AssemblyEpoch AI Benchmarking Hub–20.0%†
VIALS-AAHandshake AI–31.7%
Code2Video-AAArtificial Analysis–1021
AA-3DArtificial Analysis–954
FACTS MultimodalGoogle–51.5%
Long context
AA-LCR v1.1Artificial Analysis69.0%80.3%
Arena DocumentArena (LMArena)1440–
MLCRArtificial Analysis–26.1%
FACTS GroundingGoogle–73.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub152.3†157.6
LiveBench 2.0LiveBench–60.9%
LiveBench 1.0LiveBench–78.5%
AA Intelligence Index v5.0.1Artificial Analysis–33.6
AA Intelligence Index v4.3.2Artificial Analysis25.745.4
AA Progress Index v1Artificial Analysis281.9323.7
Vals RSI IndexVals AI–21.8%
Vals Multimodal IndexVals AI39.1%65.4%
Safety & honesty
DistressBenchScale AI SEAL–77.6%

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

  • Grok 4.20 (thinking): 1 score revision.
  • Qwen3.8 Max (max): 1 score revision.
  • Qwen3.8 Max (max): 1 score revision.
  • Qwen3.8 Max (max): 6 score revisions.
  • Qwen3.8 Max (max): 1 score revision.
  • Qwen3.8 Max (max): 1 new result (FACTS Search); 1 score revision.
  • Qwen3.8 Max (max): 1 score revision; cached input price moved from $0.22/M to $0.25/M (13% up).
  • Qwen3.8 Max (max): cached input price moved from $0.25/M to $0.22/M (11% down).
  • Qwen3.8 Max (max): 1 score revision; cached input price moved from $0.22/M to $0.25/M (13% up).
  • Qwen3.8 Max (max): cached input price moved from $0.25/M to $0.22/M (11% down).

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

Grok 4.20 vs Qwen3.8 Max: questions

Is Grok 4.20 better than Qwen3.8 Max?
Grok 4.20 (thinking) and Qwen3.8 Max (max) are neck and neck on quality (58.3 vs 58.7, within the ±5.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
Grok 4.20 vs Qwen3.8 Max: which is cheaper, and will that change?
Grok 4.20 is cheaper today: $1.56 against $3.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Grok 4.20 or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher in coding (62 vs 53 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1376 for Grok 4.20 against 1676 for Qwen3.8 Max.
Which is faster, Grok 4.20 or Qwen3.8 Max?
Grok 4.20 streams ~1.5× faster (111 vs 59 output tokens per second, Artificial Analysis 7-day median).
Is Grok 4.20 or Qwen3.8 Max better for agents and tool use?
Qwen3.8 Max scores higher in agentic tasks (59 vs 51 on the category index, where 50 is average). On Terminal-Bench 2.1 (Vals), the top shared agents & tools benchmark, it is 44.2% for Grok 4.20 against 67.4% for Qwen3.8 Max.
More questions (4)
Which is better for long documents?
Qwen3.8 Max scores higher on long-context benchmarks (65 vs 59). Grok 4.20 accepts more context: 2,000,000 against 1,000,000 tokens for Qwen3.8 Max.
Which has lower latency?
Grok 4.20 answers first: 12.1 s against 119.6 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 Grok 4.20 vs Qwen3.8 Max?
Across the 40 benchmarks both have, Grok 4.20 leads 0, Qwen3.8 Max leads 40. Every overall index that rates both puts Qwen3.8 Max ahead.
When was Qwen3.8 Max released?
Qwen3.8 Max was released on 3 Aug 2026.