Compare · ModelsLive · 2 picked · head to head
Gemma 4 26B A4B vs Qwen3 Max
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemma 4 26B A4B wins on 3/5 benchmarks
Gemma 4 26B A4B wins 3 of 5 shared benchmarks. Leads in math.
Category leads
knowledge·Qwen3 Maxgeneral·Qwen3 Maxmath·Gemma 4 26B A4B
Hype vs Reality
Attention vs performance
Gemma 4 26B A4B
#161 by perf·#7 by attention
Qwen3 Max
#160 by perf·#1 by attention
Best value
Gemma 4 26B A4B
25.9x better value than Qwen3 Max
Gemma 4 26B A4B
515.6 pts/$
$0.09/M
Qwen3 Max
19.9 pts/$
$2.34/M
Vendor risk
Who is behind the model
Google DeepMind
$3.53T·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
5 benchmarks · 2 models
Gemma 4 26B A4B Qwen3 Max
Chess Puzzles
Qwen3 Max leads by +2.9
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemma 4 26B A4B
1.1
Qwen3 Max
4.0
Dtbench
Qwen3 Max leads by +12.0
Gemma 4 26B A4B
58.2
Qwen3 Max
70.2
GPQA diamond
Gemma 4 26B A4B leads by +3.5
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemma 4 26B A4B
67.0
Qwen3 Max
63.5
Lmca
Gemma 4 26B A4B leads by +1.7
Gemma 4 26B A4B
34.9
Qwen3 Max
33.3
OTIS Mock AIME 2024-2025
Gemma 4 26B A4B leads by +8.9
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemma 4 26B A4B
82.2
Qwen3 Max
73.3
Full benchmark table
| Benchmark | Gemma 4 26B A4B | Qwen3 Max |
|---|---|---|
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 1.1 | 4.0 |
Dtbench | 58.2 | 70.2 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 67.0 | 63.5 |
Lmca | 34.9 | 33.3 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 82.2 | 73.3 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.08 | $0.10 | 262K tokens (~131 books) | $0.85 | |
| $0.78 | $3.90 | 262K tokens (~131 books) | $15.60 |
People also compared