Skip to main content

Gemma 4 26B A4B vs Gemma 4 31B

SpecGemma 4 26B A4BGemma 4 31B
Respan Index18.5 (#142)18.7 (#140)
Blended price per 1M--
Context window256K256K
Input / output per 1M$0 / $0$0 / $0
Max output--
Inputtext, visiontext, vision
Released2026-04-022026-04-02

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryGemma 4 26B A4BGemma 4 31B
Coding4.1 Gemma 4 26B A4B7.8 Gemma 4 31B
Agents13.4* Gemma 4 26B A4B12.5* Gemma 4 31B
Reasoning20.4 Gemma 4 26B A4B19.7 Gemma 4 31B
Knowledge & language15.3 Gemma 4 26B A4B11.0 Gemma 4 31B
Human preference50.0 Gemma 4 26B A4B50.4 Gemma 4 31B

* Estimated: no published scores in that category. How the Respan Index works

Benchmarks both report

Gemma 4 31B leads on 9 of the 11 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkGemma 4 26B A4BGemma 4 31B
LMArena Elo14381452
LMArena Vision (LMArena)12411265
LMArena WebDev (LMArena)13581370
Chess Puzzles (Epoch AI)6%5%
GPQA Diamond82.3%84.3%
OTIS Mock AIME 2024-2025 (Epoch AI)82.2%73.3%
WeirdML (Håvard Tveit Ihle)35.2%52.3%
AIME 202688.3%89.2%
Humanity's Last Exam (no tools)8.7%19.5%
LiveCodeBench v677.1%80%
MMLU-Pro82.6%85.2%

FAQ

Is Gemma 4 26B A4B better than Gemma 4 31B?

Gemma 4 31B scores 18.7 on the Respan Index (#140), ahead of Gemma 4 26B A4B at 18.5 (#142). Gemma 4 31B leads on 9 of the 11 benchmarks both report.

Which has the larger context window?

Both accept 256K tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.