Atlas

Compare Models

Grok 4.6 vs Qwen3.8-Max

Grok 4.6 leads by 0.3 capability points (≈95% interval −2.0 to +2.6): statistically tied · P(higher) ≈ 61%

Wins 24 of 36 shared benchmarks (12 lost)

Capability vs. relative cost
Grok 4.6Qwen3.8-MaxClosed weightsOpen weights

Each line runs through one model’s priced reasoning settings, cheapest to most expensive; grey points are other models’ priced settings in the same range. Results whose reasoning effort the source didn’t report are left out wherever a model has named effort levels. Capability and relative cost are the published estimates (relative cost is cost per task as a multiple of GPT-5 (high)); hover a setting, or use the arrow keys, to set it beside the other model’s same setting.

Grok 4.6vsQwen3.8-Maxcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins24 – 12 · 36 shared

Comparing matching setups only (284 benchmarks excluded) · suites counted once
← Grok 4.6Qwen3.8-Max →

Grok 4.6vsQwen3.8-Maxwins by categoryData Nov 15, 2026Data: Artificial Analysis, Conceptual Reasoning Benchmark Authors, Context Arena and 13 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Qwen3.8-Max minus Grok 4.6.

MRCR v2 (8-needle, Context Arena) - 128K pointwiseGeneral QA52.692.3+39.7
BabaIsBench Stage 0: The IntroGames96.058.0−38.0
MRCR v2 (8-needle, Context Arena) - 128K AUCGeneral QA73.496.1+22.7
YC-BenchGames15682412095838+18.4
AutomationBench-AA 1.0.6Agents67.049.2−17.8
GameBench 2Games68.751.4−17.3
AA-Omniscience IndexIndexes30.53.4−13.5
AA-Briefcase EloAgents15771388−11.0
SCI-AGENT-EXAMScience52.462.7+10.3
BuseyBench SVGMultimodal4.65.4+8.1
Medical Long Context Reasoning (MLCR-AA)Health & Medicine12.219.4+7.2
Terminal-Bench 2.1Code88.481.3−7.1
GDPval-AA v2Professional Work17531630−6.0
DTBenchGeneral QA97.392.0−5.3
FrontierChallenge (Pass Rate)Science20.615.5−5.2
RSI-Exam 0.1 (Full 88)Agents0.40.3−4.5
Conceptual Reasoning IndexIndexes62.559.4−4.3
Artificial Analysis Intelligence Index v4.3.2Indexes44.340.2−4.2
Artificial Analysis Intelligence Index v4.3Indexes44.440.3−4.1
Humanity's Last ExamGeneral QA39.743.6+3.9
SciCode 1.0.1 (Artificial Analysis)Code56.553.2−3.2
xbench-ScienceQAScience70.473.4+3.0
AA-LCR v1.1Search81.078.3−2.7
Terminal-Bench 4.0 (Artificial Analysis)Code21.218.7−2.5
GDP.pdf (Artificial Analysis, All-pass)Professional Work17.820.2+2.4
GPQA Diamond (Artificial Analysis avg@5)General QA94.992.7−2.2
LMCAGeneral QA48.546.2−2.2
GDPval-AA v2.1Professional Work16581611−2.1
MMMU ProMultimodal80.582.3+1.8
Artificial Analysis Progress Index v1Indexes322320−0.7
DeepsecBench — Revalidated V1Code16.215.5−0.6
DeepsecBench — Revalidated V1 (233 issues)Code16.115.5−0.6
𝜏³-BankingAgents50.751.3+0.6
Epoch Capabilities Index (ECI)Indexes158157−0.4
CritPt (avg@5)Science19.720.0+0.3
ACCoRDGeneral QA71.671.4−0.2

Grok 4.6vsQwen3.8-Maxnormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Conceptual Reasoning Benchmark Authors, Context Arena and 13 othersVillage LLM Atlasatlas.village.dev

Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…