Skip to main content

Gemma 4 26B A4B vs Qwen3 235B-A22B Thinking 2507

SpecGemma 4 26B A4BQwen3 235B-A22B Thinking 2507
Respan Index18.5 (#142)18.5 (#142)
Blended price per 1M-$1.21
Context window256K262K
Input / output per 1M$0 / $0$0.45 / $3.5
Max output--
Inputtext, visiontext
Released2026-04-022025-07-25

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 A4BQwen3 235B-A22B Thinking 2507
Coding4.1 Gemma 4 26B A4B5.2 Qwen3 235B-A22B Thinking 2507
Agents13.4* Gemma 4 26B A4B4.1 Qwen3 235B-A22B Thinking 2507
Reasoning20.4 Gemma 4 26B A4B22.1 Qwen3 235B-A22B Thinking 2507
Knowledge & language15.3 Gemma 4 26B A4B38.1 Qwen3 235B-A22B Thinking 2507
Human preference50.0 Gemma 4 26B A4B45.4 Qwen3 235B-A22B Thinking 2507

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

Benchmarks both report

Qwen3 235B-A22B Thinking 2507 leads on 4 of the 6 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkGemma 4 26B A4BQwen3 235B-A22B Thinking 2507
LMArena Elo14381400
Chess Puzzles (Epoch AI)6%12%
GPQA Diamond82.3%80.1%
OTIS Mock AIME 2024-2025 (Epoch AI)82.2%86.7%
WeirdML (Håvard Tveit Ihle)35.2%41%
MMLU-Pro82.6%84.4%

FAQ

Is Gemma 4 26B A4B better than Qwen3 235B-A22B Thinking 2507?

Gemma 4 26B A4B and Qwen3 235B-A22B Thinking 2507 tie on the Respan Index at 18.5. Qwen3 235B-A22B Thinking 2507 leads on 4 of the 6 benchmarks both report.

Which has the larger context window?

Gemma 4 26B A4B accepts 256K tokens and Qwen3 235B-A22B Thinking 2507 262K tokens.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.