Gemma 4 31B
- Input / output
- $0.10 / $0.34
- Providers
- 15
- Spread
- 4.8×
- 7 days
- 0%
- 30 days
- 0%
- Context
- 256K
Providers
| Provider | Input | Output | Blended | vs ref | Context | Uptime 1d | Price since |
|---|---|---|---|---|---|---|---|
CoreWeaveCheapest fp4 | $0.10 | $0.34 | $0.16 | −5.5% | |||
| $0.10 | $0.34 | $0.16 | −5.5% | ||||
| fp4 | $0.125 | $0.34 | $0.179 | +5.5% | |||
| fp4 | $0.12 | $0.36 | $0.18 | +6.3% | |||
| fp4 | $0.12 | $0.37 | $0.183 | +7.7% | |||
DeepInfrafp8/flex fp8 | $0.16 | $0.32 | $0.20 | +18% | |||
| bf16 | $0.14 | $0.40 | $0.205 | +21% | |||
| $0.14 | $0.40 | $0.205 | +21% | ||||
| bf16 | $0.14 | $0.40 | $0.205 | +21% | |||
| fp8 | $0.15 | $0.40 | $0.213 | +25% | |||
| fp8 | $0.185 | $0.31 | $0.216 | +28% | |||
DeepInfraturbo fp4 | $0.135 | $0.51 | $0.229 | +35% | |||
| fp8 | $0.20 | $0.40 | $0.25 | +48% | |||
| $0.34 | $0.10 | $0.28 | +65% | ||||
DeepInfrafp8/priority fp8 | $0.30 | $0.60 | $0.375 | +121% | |||
| $0.38 | $1.15 | $0.573 | +238% | ||||
| fp4 | $0.75 | $1.00 | $0.813 | +380% | |||
| fp8 | $0.75 | $1.00 | $0.813 | +380% |
- CoreWeaveCheapest$0.16$0.10 in · $0.34 out−5.5% vs ref256K ctx100% uptimefp4
- $0.16$0.10 in · $0.34 out−5.5% vs ref256K ctx99.9% uptime
- $0.179$0.125 in · $0.34 out+5.5% vs ref256K ctx99.9% uptimefp4
- $0.18$0.12 in · $0.36 out+6.3% vs ref250K ctx100% uptimefp4
- $0.183$0.12 in · $0.37 out+7.7% vs ref128K ctx99.9% uptimefp4
- fp8/flex$0.20$0.16 in · $0.32 out+18% vs ref256K ctx99.0% uptimefp8
- $0.205$0.14 in · $0.40 out+21% vs ref256K ctx100% uptimebf16
- $0.205$0.14 in · $0.40 out+21% vs ref256K ctx100% uptime
- $0.205$0.14 in · $0.40 out+21% vs ref256K ctx97.8% uptimebf16
- $0.213$0.15 in · $0.40 out+25% vs ref256K ctx98.5% uptimefp8
- $0.216$0.185 in · $0.31 out+28% vs ref256K ctx99.4% uptimefp8
- turbo$0.229$0.135 in · $0.51 out+35% vs ref256K ctx100% uptimefp4
- $0.25$0.20 in · $0.40 out+48% vs ref256K ctx100% uptimefp8
- $0.28$0.34 in · $0.10 out+65% vs ref128K ctx100% uptime
- fp8/priority$0.375$0.30 in · $0.60 out+121% vs ref256K ctx99.6% uptimefp8
- $0.573$0.38 in · $1.15 out+238% vs ref128K ctx92.4% uptime
- $0.813$0.75 in · $1.00 out+380% vs ref256K ctx100% uptimefp4
- $0.813$0.75 in · $1.00 out+380% vs ref256K ctx100% uptimefp8
Blended = (3 × input + output) / 4. Ref $0.169 = median of cheapest 20%.
Price history
USD per 1M tokens · each step is a price change
Prices in USD / 1M tokens.
Other pricing
| Offered by | Cache read |
|---|---|
| CoreWeave, KT +1 | $0.10 |
| NextBit, DeepInfra (turbo) | $0.075 |
| Venice | $0.09 |
| Chutes | $0.012 |
| Crusoe | $0.14 |
| Parasail | $0.06 |
| PrimeIntellect | $0.092 |
| ModelRun | $0.20 |
| SiliconFlow | $0.25 |
Changes
- PriceCache read price,$0.076→$0.092+21.1%
- PriceOutput price,$0.255→$0.31+21.6%
- PriceInput price,$0.152→$0.185+21.7%
- PriceCache read price,$0.10→$0.076−24.0%
- PriceOutput price,$0.335→$0.255−23.9%
- PriceInput price,$0.20→$0.152−24.0%
- PriceCache read price,$0.076→$0.10+31.6%
- PriceOutput price,$0.254→$0.335+31.9%
- PriceInput price,$0.152→$0.20+31.6%
- PriceCache read price,$0.10→$0.076−24.0%
- PriceOutput price,$0.335→$0.254−24.2%
- PriceInput price,$0.20→$0.152−24.0%
- PriceCache read price,$0.06→$0.075+25.0%
- PriceOutput price,$0.40→$0.34−15.0%
- PriceInput price,$0.15→$0.125−16.7%
About
The cheapest provider for Gemma 4 31B is CoreWeave at $0.16 per 1M tokens (blended), charging $0.10 per 1M input and $0.34 per 1M output tokens. 15 providers serve it; the most expensive charges 4.8× the typical price. Context window: 256K tokens. Prices tracked since Oct 3, 2026.
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...
- Released
- Apr 2, 2026
- First tracked
- Oct 3, 2026
- Canonical slug
- google/gemma-4-31b-it-20260402
- Weights
- google/gemma-4-31B-it