⌕Is ___ SOTA?
isitsota / models / Google DeepMind / gemma-4-26b-a4b
VERIFIEDOct 142026
Is Gemma 4 26B A4b state of the art?
Yes.Yes on General LLM (≤32B); Reasoning (≤32B). Not the overall leader.

As of Oct 14, 2026, Gemma 4 26B A4b from Google DeepMind. Yes on General LLM (≤32B); Reasoning (≤32B). Not the overall leader. Ranked #38 in Math.

Computed automaticallyas of Oct 14, 2026methodology v1.0confidence: low
General LLMNot SOTA · #73#73 · 0dGeneral LLM · le-32bCo-SOTA#3 · 6dReasoningNot SOTA · #70#70 · 0dReasoning · le-32bCo-SOTA#3 · 6dMathNot SOTA · #38#38 · 0dCodingNot SOTA · #75#75 · 0dCoding › Web devNot SOTA · #81#81 · 0dLong contextNot SOTA · #72#72 · 0dLong context · le-32bCo-SOTA#2 · 6dVisionNot SOTA · #50#50 · 0d
Alert me if this changesGet badgeCompare with….mdJSON
View as of point-in-time crowns
Why

The ruling, in plain English

Ranked #73 in General LLM.

Receipts

Evidence

Scores come from the latest ingested snapshot
SourceScoreDateTypeReceipt
LMArena Text (style-controlled, overall)
gemma-4-26b-a4b
1437.6181934274218 1430.2–1445.12026-10-08INDEP.snap_368d64825bf4
LMArena Text (style-controlled, hard prompts)
gemma-4-26b-a4b
1461.1043545557072 1451.3–1470.92026-10-08INDEP.snap_368d64825bf4
LMArena Text (style-controlled, math)
gemma-4-26b-a4b
1468.9623942746111 1441.4–1496.52026-10-08INDEP.snap_368d64825bf4
LMArena Text (style-controlled, coding)
gemma-4-26b-a4b
1482.7275259778385 1468.0–1497.52026-10-08INDEP.snap_368d64825bf4
LMArena WebDev
gemma-4-26b-a4b
1358.9928359404478 1342.1–1375.92026-10-08INDEP.snap_6949d6fb2488
LMArena Text (style-controlled, longer query)
gemma-4-26b-a4b
1446.503140882518 1432.7–1460.32026-10-08INDEP.snap_368d64825bf4
LMArena Vision (style-controlled)
gemma-4-26b-a4b
1241.9593762444456 1235.5–1248.42026-10-09INDEP.snap_44d0155ee751
LMArena Text factuality
gemma-4-26b-a4b
1444.8655770283908 1438.0–1451.82026-10-08INDEP.snap_f051faf18005
Reign history

Where it sits in the succession

Full timeline →

No closed or open crown interval includes this model.

Honest about the gaps

Where it is not the crown

CategoryThis modelWhy
General LLMNot SOTA · #73Ranked #73 in General LLM.
ReasoningNot SOTA · #70Ranked #70 in Reasoning.
MathNot SOTA · #38Ranked #38 in Math.
CodingNot SOTA · #75Ranked #75 in Coding.
Coding › Web devNot SOTA · #81Ranked #81 in Coding › Web dev.
Long contextNot SOTA · #72Ranked #72 in Long context.
VisionNot SOTA · #50Ranked #50 in Vision.
FactualityNot SOTA · #60Ranked #60 in Factuality.