GoogleGemma 4 12B (Non-reasoning)
- Released
- Jun 3, 2026
- Tracked since
- 2026-07-23
Intelligence index9.4Ranks 392 of 664 models with this metric
Coding index—Not tested
Agentic index—Not tested
Capability profile
5 axes vs. site averageGemma 4 12B (Non-reasoning)
Site average
Some dimensions have no data yet — marked on the chart
Pricing
per 1M tokens- Input
- $0.100
- Output
- $0.300
Performance
- Output throughput
- 123 tokens/sec
- Time to first token
- 2.42s
- First answer token
- 2.42s
- End-to-end response
- 6.49s
History
1Intelligence index
13.2→9.4
Output
Unchanged across 62 days of records
Output throughput
98→123
Cheaper at this level
Models with a comparable intelligence index and a lower output price- Granite 4.2 3B
IBMIntelligence 9.1$0.120−60% - Qwen3.5 4B (Reasoning)
AlibabaIntelligence 13.1$0.150−50% - Qwen3.5 4B (Non-reasoning)AlibabaIntelligence 10.8$0.150−50%
- gpt-oss-20b (high)
OpenAIIntelligence 9$0.180−40% - DeepSeek V4 Flash 0420 (Non-reasoning)
DeepSeekIntelligence 18.9$0.200−33%