GoogleOpen weightsReleased Apr 2, 2026

Gemma 4 31B pricing, benchmarks and cost analysis

Compare Google guide

Intelligence#61 / 66

143.0

Epoch Capabilities Index (ECI)range 140.3 to 145.8

Speed

No speed data

Throughput not measured for this model

Cost#13 / 101

$137 / mo

In $0.200Out $0.400 per 1MCache discount 90%

Context#79 / 101

262K

Context window, tokens

Ranks are among the 101 models we track (intelligence = Epoch Capabilities Index by Epoch AI, CC BY 4.0; intelligence and speed: among those with data; cost: #1 is cheapest). Monthly cost uses the default workload of 2,000 input and 500 output tokens, 10,000 requests a day, 40% cached.

Comparison summary

On the default workload (2,000 input and 500 output tokens, 10,000 requests a day, 40% cached), Gemma 4 31B comes to $137 a month, 84% below the $873 median of the 101 models we track. It ranks #13 of 101 on cost, where #1 is the cheapest.

It scores 143 on the Epoch Capabilities Index (#61 of 66). Among the 9 other models within 2 points, 2 cost less, and it costs 43% less than their median ($241). The cheapest stable-priced model that scores at least as high is Qwen3.7 Flash (144.1) at $31.74 a month, 77% less.

Cached input is 90% cheaper than fresh input, in line with the 90% median. The weights are open, so other hosts can serve it and prices may differ from the provider's own.