Model specification · Google
Gemma 4 26B A4B
google/gemma-4-26b-a4b-it
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
SpecificationsReported values
- Context window
- 262K
- Maximum output
- 16K
- Architecture
- text+image+video->text
- Tokenizer
- Gemma
- Knowledge cutoff
- —
- Moderated
- No
- Input modalities
- image text video
- Output modalities
- text
Full pricingUSD where reported
| Price | Standard | free |
|---|---|---|
| Input | $0.070 / M | free / M |
| Output | $0.340 / M | free / M |
Benchmark indexesReported indexes
- Intelligence
- 25.7
- Coding
- 39.3
- Agentic
- 11
- Design Arena Elo
- —
Example workload costsStandard pricing only
| Workload | Input / output | Cache share | Estimated cost |
|---|---|---|---|
| Quick check | 10K / 1K × 1 | 0% | $0.00104 |
| Long analysis | 100K / 4K × 1 | 50% | $0.00836 |
| Workday | 20K / 2K × 1,000 | 30% | $2.08 |
More from GoogleAll Google models
| Model | Context | Input / M | Output / M | Intel. |
|---|---|---|---|---|
| Gemini 3.5 Flash | 1.0M | $1.50 | $9.00 | 50.2 |
| Gemini 3.6 Flash | 1.0M | $1.50 | $7.50 | 50.1 |
| Gemini 3.1 Pro Preview | 1.0M | $2.00 | $12.00 | 46.5 |
| Gemini 3.5 Flash Lite | 1.0M | $0.300 | $2.50 | 36.5 |
| Gemma 4 31B | 262K | $0.100 | $0.340 | 29.4 |
| Gemini 2.5 Pro | 1.0M | $1.25 | $10.00 | 25.8 |
| Gemini 3.1 Flash Lite Preview | 1.0M | $0.250 | $1.50 | 25 |
| Gemma 3 27B | 262K | $0.080 | $0.450 | 7.4 |
| Gemma 3 12B | 131K | $0.050 | $0.150 | 5.5 |