Gemma 4 31B
by Google · google/gemma-4-31b-it
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...
- Input / 1M tokens
- $0.20
- Output / 1M tokens
- $0.40
- Context
- 262K
Specs
| Input price | $0.20 / 1M tokens |
|---|---|
| Output price | $0.40 / 1M tokens |
| Context window | 262K tokens |
| Max output | 16.4K tokens |
| Reads | image, text, video |
| Writes | text |
| Tool calling | Yes |
| Reasoning | Yes |
| Listed | Apr 2, 2026 |
List prices in USD, updated Oct 30, 2026. Providers may charge differently.
Providers
16 providers serve this model, as of Oct 30, 2026, 18:12 UTC.
| Provider | Quantization | Context | Max output | Input / 1M | Output / 1M | Cache read / 1M | Uptime (30 min) | Latency (median) | Throughput (median) | Supports | Status |
|---|---|---|---|---|---|---|---|---|---|---|---|
| CoreWeave | fp4 | 262K | 236K | $0.10 | $0.34 | $0.10 | 100% | 1.01 s | 32 tok/s | Normal | |
| DeepInfra | fp8 | 262K | 236K | $0.16 | $0.32 | — | — | 3.51 s | 12 tok/s | Normal | |
| Venice | fp4 | 256K | 8.19K | $0.12 | $0.36 | $0.09 | 100% | 647 ms | 30 tok/s | Normal | |
| Chutes | fp4 | 131K | 65.5K | $0.12 | $0.37 | $0.012 | 60% | 5.20 s | 19 tok/s | Code -5 | |
| Crusoe | bf16 | 262K | 262K | $0.14 | $0.40 | $0.14 | 100% | 320 ms | 83 tok/s | Normal | |
| DekaLLM | — | 262K | 236K | $0.14 | $0.40 | $0.05 | 100% | 1.63 s | 21 tok/s | Normal | |
| Friendli | — | 262K | 8.19K | $0.14 | $0.40 | — | 100% | 851 ms | 101 tok/s | Normal | |
| Novita | bf16 | 262K | 131K | $0.14 | $0.40 | — | 90% | 733 ms | 28 tok/s | Normal | |
| Parasail | fp8 | 262K | 236K | $0.15 | $0.40 | $0.06 | 100% | 1.53 s | 19 tok/s | Normal | |
| DeepInfra | fp8 | 262K | 16.4K | $0.20 | $0.40 | — | 96.6% | 765 ms | 21 tok/s | Normal | |
| DeepInfra | fp4 | 262K | 16.4K | $0.135 | $0.51 | $0.075 | 100% | 599 ms | 36 tok/s | Normal | |
| DeepInfra | fp8 | 262K | 236K | $0.30 | $0.60 | — | 100% | 1.70 s | 12 tok/s | Normal | |
| Io Net | — | 262K | 16.4K | $0.361 | $1.09 | $0.18 | 100% | 1.11 s | 23 tok/s | Normal | |
| SambaNova | — | 131K | 118K | $0.38 | $1.15 | — | 92.9% | 2.46 s | 147 tok/s | Code -2 | |
| ModelRun | fp4 | 262K | 236K | $0.75 | $1.00 | $0.20 | 100% | 181 ms | 182 tok/s | Normal | |
| SiliconFlow | fp8 | 262K | 236K | $0.75 | $1.00 | $0.25 | 100% | 1.87 s | 9 tok/s | Normal |
"Cheapest" is the lowest input plus output price; "Fastest" the highest median throughput. Columns empty for every provider are not shown.
Usage
#43
by tokens, Oct 29, 2026
78.2B
tokens that day · 7 days −20%
Data as of 2026-10-30T18:12:03.879Z.All rankings
Related models
- Google
Gemini 3.8 Flash
- Input
- $0.825
- Output
- $4.13
- Context
- 1.05M
- Google
Gemini 4 Argon
- Input
- $2.00
- Output
- $10.00
- Context
- 1.05M
- Google
Gemini 3.1 Flash Lite
- Input
- $0.25
- Output
- $1.50
- Context
- 1.05M
- Google
Gemini 3 Flash Preview
- Input
- $0.50
- Output
- $3.00
- Context
- 1.05M
- Google
Gemma 4 26B A4B
- Input
- $0.09
- Output
- $0.30
- Context
- 262K
- Google
Gemini 4 Argon (batch)
- Input
- $1.00
- Output
- $5.00
- Context
- 1.05M