Updated March 26, 2026· Based on independent benchmark data
Google: Gemma 4 31B is 13.5x cheaper at $0.13/1M tokens vs $1.75/1M.
| Metric | Google: Gemma 4 31B | GPT-5.3 Codex (xhigh) |
|---|---|---|
| Intelligence Score | N/A | 54.0 |
| Coding Score | N/A | 53.1 |
| Math Score | N/A | N/A |
| Speed (tok/s) | N/A | 72 tok/s |
| Latency (TTFT) | N/A | 83.91s |
| Input Price / 1M tokens | $0.13 | $1.75 |
| Output Price / 1M tokens | $0.38 | $14 |
| Context Window | 262K | N/A |
| Max Output Tokens | 16K | N/A |
| Input Modalities | Image + Text + Video | Text |
| Output Modalities | Text | Text |
| Free Tier | No | No |
Google: Gemma 4 31B is more affordable at $0.13/1M input tokens ($0.38/1M output), while GPT-5.3 Codex (xhigh) costs $1.75/1M input ($14/1M output). That makes GPT-5.3 Codex (xhigh) 13.5x more expensive per token, which can add up significantly at scale. For a typical workload of 100 requests per day at 2,000 tokens each, Google: Gemma 4 31B would cost approximately $0.78/month vs $10.50/month for GPT-5.3 Codex (xhigh) in input costs alone.
Choose Google: Gemma 4 31B when you need lower cost.
Google: Gemma 4 31B is cheaper at $0.13/1M input tokens vs $1.75/1M for GPT-5.3 Codex (xhigh).
Yes, Google: Gemma 4 31B supports image input. GPT-5.3 Codex (xhigh) does not support image input.
Data last synced: March 26, 2026