Pricing
| Provider | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|
| Sail Research | $0.14 | $0.07 | $0.40 |
| Chutes | $0.12 | $0.012 | $0.37 |
USD, as of November 5, 2026. Live prices
Gemma 4 31B NVFP4 API
- 262KContext
- 262KMax output
- $0.12/ $0.37From, per 1M in / out
- 2Providers
Providers
Cost
| Workload | Sail Research | Chutes |
|---|---|---|
| 1,000 chat replies2K in, 500 out each | $0.48 | $0.43 |
| 1,000 agent tasks50K in, 5K out each | $9.00 | $7.85 |
| Monthly batch job500M in, 50M out | $90.00 | $78.50 |
Gemma 4 versions
| Model | Released | Context | Lowest input | Lowest output | Providers |
|---|---|---|---|---|---|
| Gemma 4 12Bgoogle/gemma-4-12b-it | May 23, 2026 | 16.4K | $0.30 | $2.00 | 1 |
| Gemma 4 31Bgoogle/gemma-4-31b-it | April 2, 2026 | 262K | $0.12 | $0.35 | 4 |
| Gemma 4 31B NVFP4This pagegoogle/gemma-4-31b-it-nvfp4 | April 2, 2026 | 262K | $0.12 | $0.37 | 2 |
Lowest listed price, USD per 1M tokens, as of November 5, 2026.
Quickstart
curl https://api.routerplus.com/v1/chat/completions \
-H "Authorization: Bearer $ROUTERPLUS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google/gemma-4-31b-it-nvfp4",
"messages": [{ "role": "user", "content": "Hello" }]
}'FAQ
How much does the Gemma 4 31B NVFP4 API cost?
On RouterPlus, routing for Gemma 4 31B NVFP4 starts at Sail Research: $0.14 per million input tokens and $0.40 per million output tokens. The lowest listed prices are at Chutes: $0.12 input and $0.37 output. Name a provider with provider.only and you pay its price. Prices as of November 5, 2026.
Which providers serve Gemma 4 31B NVFP4?
2 providers serve Gemma 4 31B NVFP4 on RouterPlus: Sail Research and Chutes. By default the router tries them in routing order and fails over before the first byte, never mid-answer.
What is the Gemma 4 31B NVFP4 model ID?
Use "google/gemma-4-31b-it-nvfp4" as the model with the OpenAI-compatible endpoint at https://api.routerplus.com/v1. The same key also accepts the Anthropic API format.
What is the Gemma 4 31B NVFP4 context window?
262K tokens of context and up to 262K output tokens, per the RouterPlus catalog. A provider can list a lower limit; the console shows each one.
Is Gemma 4 31B NVFP4 open weights?
Yes. Google publishes Gemma 4 31B NVFP4 as an open-weight model, which is why several independent providers can host it.
Do any Gemma 4 31B NVFP4 providers retain prompts?
None of the 2 providers serving Gemma 4 31B NVFP4 on RouterPlus list prompt retention.
Can I reserve capacity for Gemma 4 31B NVFP4?
Yes. Tell us the model, the throughput you need and the region with "I need inference", and RouterPlus lines up reserved capacity for it.