Googleopen weightsGemma 4 31B pricing
$0.20 per 1M input tokens and $0.40 per 1M output tokens. In real work that is 0.07¢ per chat answer and 0.30¢ to summarize a 20-page PDF.
What Gemma 4 31B costs per task
Token counts estimated from a measured model of the same lab (Gemini 3.8 Flash), rescaled by tokenizer. Reasoning tokens are billed as output.
| Task | Cost | Tokens used | Price rank | Basis |
|---|---|---|---|---|
| Answer a chat question | 0.07¢$0.66 / 1,000 | 35 in · 788 out · 843 reasoning | 74 of 336 | estimated |
| Reply to a customer email | 0.03¢$0.28 / 1,000 | 271 in · 128 out · 424 reasoning | 75 of 336 | estimated |
| Write a 1,500-word blog post | 0.14¢$1.44 / 1,000 | 52 in · 2,621 out · 964 reasoning | 43 of 336 | estimated |
| Write 10 product descriptions | 0.12¢$1.24 / 1,000 | 293 in · 1,127 out · 1,817 reasoning | 37 of 336 | estimated |
| Translate 1,000 words | 0.20¢$2.05 / 1,000 | 1,308 in · 1,297 out · 3,167 reasoning | 104 of 336 | estimated |
| Proofread one page (500 words) | 0.06¢$0.58 / 1,000 | 706 in · 653 out · 443 reasoning | 65 of 336 | estimated |
| Summarize a 20-page PDF | 0.30¢$3.05 / 1,000 | 12,741 in · 535 out · 713 reasoning | 87 of 336 | estimated |
| Extract data from an invoice | 0.07¢$0.68 / 1,000 | 452 in · 539 out · 943 reasoning | 73 of 336 | estimated |
| Analyze a 200-row spreadsheet | 0.45¢$4.51 / 1,000 | 6,304 in · 1,018 out · 7,100 reasoning | 63 of 336 | estimated |
| Answer from 5 retrieved documents (RAG) | 0.16¢$1.59 / 1,000 | 3,608 in · 669 out · 1,499 reasoning | 88 of 336 | estimated |
| Classify 100 support tickets | 0.29¢$2.93 / 1,000 | 1,397 in · 571 out · 6,059 reasoning | 108 of 336 | estimated |
| Review 500 lines of code | 0.19¢$1.87 / 1,000 | 5,207 in · 2,074 out | 18 of 336 | estimated |
| Write a SQL query | 0.07¢$0.66 / 1,000 | 135 in · 546 out · 1,043 reasoning | 66 of 336 | estimated |
| Run a 50-step coding agent | 46¢$464 / 1,000 | 2,194,987 in · 14,715 out · 47,150 reasoning | 164 of 336 | modeled |
Price at every provider
List prices per 1M tokens (input / output) at each host serving Gemma 4 31B through OpenRouter today. Hosts can differ in quantization, context and speed.
| Provider | Input / output | Cached input | Context | Uptime |
|---|---|---|---|---|
 DeepInfrafp8 | $0.16 / $0.32 | - | 262K | 100% |
| C CoreWeavefp4 | $0.10 / $0.34 | $0.10 | 262K | 100% |
| V Venicefp4 | $0.12 / $0.36 | $0.09 | 256K | 100% |
 Chutesfp4 | $0.12 / $0.37 | $0.012 | 131K | 100% |
| C Crusoebf16 | $0.14 / $0.40 | $0.14 | 262K | 100% |
| F Friendliunknown | $0.14 / $0.40 | - | 262K | 100% |
 Novitabf16 | $0.14 / $0.40 | - | 262K | 77% |
 Parasailfp8 | $0.15 / $0.40 | $0.06 | 262K | 100% |
DeepInfrafp8 | $0.20 / $0.40 | - | 262K | 100% |
DeepInfrafp4 | $0.14 / $0.51 | $0.075 | 262K | 100% |
DeepInfrafp8 | $0.30 / $0.60 | - | 262K | 100% |
| I Io Netunknown | $0.36 / $1.09 | $0.18 | 262K | 100% |
| M ModelRunfp4 | $0.75 / $1.00 | $0.20 | 262K | 100% |
 SiliconFlowfp8 | $0.75 / $1.00 | $0.25 | 262K | 100% |
 SambaNovaunknown | $0.38 / $1.15 | - | 131K | 100% |
Price history
First seen in our price history on Apr 7, 2026 at $0.14 input / $0.40 output. Since then the output price moved from $0.40 to $0.40 per 1M tokens (10 changes).
FAQ
How much does Gemma 4 31B cost?
Gemma 4 31B costs $0.20 per 1 million input tokens and $0.40 per 1 million output tokens (list price on Oct 18, 2026). In practice one chat answer costs about 0.07¢ and summarizing a 20-page PDF about 0.30¢.
Where is Gemma 4 31B cheapest?
Among 15 providers we track, the lowest list price today is at DeepInfra: $0.16 input and $0.32 output per 1M tokens.
Is the Gemma 4 31B cost per task measured or estimated?
Estimated: Gemma 4 31B was not run directly. Tokens come from Gemini 3.8 Flash and are rescaled by the tokenizer of Google. Real usage can differ.