Gemma 4 31B
Google · Open weights · Released Apr 2, 2026 · google/gemma-4-31b-it
Best $0.10 in / $0.335 out via CoreWeave · fp4
262K contextVisionToolsReasoningStructured outputPrompt cachingOpen weights
Available from 18 providersUSD / MTOK · PRICES 15 MINUTES AGO · SPEED 8 MINUTES AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
| $0.10 | $0.34 | $0.10 | 262K | fp4 | 1.4s | 61 tok/s | 5/5 | 99.9% | |
| KTKT | $0.10 | $0.34 | $0.10 | 262K | — | — | — | — | 100.0% |
| $0.12 | $0.37 | $0.012 | 131K | fp4 | ✕ | ✕ | 5/5 | 99.3% | |
| $0.12 | $0.36 | $0.09 | 256K | fp4 | 2.1s | 86 tok/s | 5/5 | 99.9% | |
| $0.125 | $0.34 | $0.075 | 262K | fp4 | 1.4s | 138 tok/s | — | 100.0% | |
| $0.135 | $0.51 | $0.075 | 262K | fp4 | 0.56s | 30 tok/s | 5/5 | 98.6% | |
| $0.14 | $0.40 | $0.14 | 262K | bf16 | ✕ | ✕ | — | 100.0% | |
| $0.14 | $0.40 | — | 262K | — | 0.83s | 179 tok/s | — | 100.0% | |
| $0.14 | $0.40 | — | 262K | bf16 | ✕ | ✕ | — | 99.9% | |
| $0.15 | $0.40 | $0.06 | 262K | fp8 | 1.6s | 62 tok/s | — | 100.0% | |
| $0.16 | $0.32 | — | 262K | fp8 | — | — | — | 100.0% | |
| $0.20 | $0.40 | — | 262K | fp8 | — | — | — | 100.0% | |
| $0.20 | $0.335 | $0.10 | 262K | fp8 | 1.4s | 45 tok/s | — | 99.8% | |
| $0.30 | $0.60 | — | 262K | fp8 | — | — | — | 99.9% | |
| $0.34 | $0.10 | $0.10 | 131K | — | 0.49s | 106 tok/s | — | 100.0% | |
| $0.38 | $1.15 | — | 131K | — | 3.6s | 343 tok/s | — | 95.4% | |
| $0.75 | $1.00 | $0.20 | 262K | fp4 | — | — | — | 100.0% | |
| $0.75 | $1.00 | $0.25 | 262K | fp8 | 3.2s | 64 tok/s | 5/5 | 100.0% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 8 minutes ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Input price history · best listed price, observed
↑ 11% SINCE SEP 27, 2026
$0.09 · SEP 27, 2026$0.10 · TODAY
Pricing variants · USD / MTok in / out
STANDARD
$0.10 / $0.34
per MTok in / out
FREE
Free / Free
per MTok in / out
CACHE READ $0.10
Recent changesRSS ↗
Nov 19, 2026price increased · output0.226 → 0.335
Nov 18, 2026price increased · input0.068 → 0.1
Nov 18, 2026price increased · output0.115 → 0.226
Nov 17, 2026endpoint added · modelrun18 → 18
Nov 17, 2026endpoint removed · dekallm18 → 18
Nov 12, 2026endpoint removed · io-net19 → 18
Nov 11, 2026endpoint added · kt19 → 19
Nov 11, 2026endpoint removed · modelrun19 → 19
About
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...
Specifications
Context window262,144
Max output16,384
Modalities inimage, text, video
Modalities outtext
LicenseOpen weights
ReleasedApr 2, 2026
TokenizerGemma
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/google/gemma-4-31b-it)Sources
OpenRouter API15 minutes ago
AGGREGATORHugging Face15 minutes ago
COMMUNITYModelIndex benchmarks20 days ago
MEASUREDVerified 15 minutes ago · list prices, not negotiated rates.