LiveBench Overall ↗2.0 · 2026-10-14 (dataset 2026-09-02)
41.53%Effort: max · single run · equal mean of 7 suites
googleListed Mar 11, 2026
Google: Gemma 4 31B
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...
Context window262.1K
Max output16.4K
Parameters31B
Provider endpoints20
Benchmark results8
Measured performance
All leaderboards ↗Benchmark results
LiveBench Design ↗2.0 · 2026-10-14 (dataset 2026-09-02)
58.4%Effort: max · single run
LiveBench Research ↗2.0 · 2026-10-14 (dataset 2026-09-02)
44.4%Effort: max · single run
LiveBench Browser Use ↗2.0 · 2026-10-14 (dataset 2026-09-02)
4%Effort: max · single run
LiveBench Coding ↗2.0 · 2026-10-14 (dataset 2026-09-02)
49%Effort: max · single run
LiveBench Agentic ↗2.0 · 2026-10-14 (dataset 2026-09-02)
13.3%Effort: max · single run
LiveBench Data Analysis ↗2.0 · 2026-10-14 (dataset 2026-09-02)
53.3%Effort: max · single run
LiveBench Reasoning ↗2.0 · 2026-10-14 (dataset 2026-09-02)
68.3%Effort: max · single run
Access & cost
20 quotesPrices across providers
Your token mix
USD · listed token ratesOpenRouter
Catalog quote- Input
- $0.20
- Output
- $0.40
novita
via Hugging Face- Input
- $0.14
- Output
- $0.40
featherless-ai
via Hugging Face- Input
- —
- Output
- —
deepinfra
via Hugging Face- Input
- $0.20
- Output
- $0.40
- Input
- $0.20
- Output
- $0.40
- Input
- $0.10
- Output
- $0.34
- Cached input
- $0.10
- Input
- $0.12
- Output
- $0.36
- Cached input
- $0.09
- Input
- $0.12
- Output
- $0.37
- Cached input
- $0.012
- Input
- $0.135
- Output
- $0.51
- Cached input
- $0.075
- Input
- $0.14
- Output
- $0.40
- Cached input
- $0.14
- Input
- $0.14
- Output
- $0.40
- Input
- $0.14
- Output
- $0.40
Endpoint degraded (status -2)
- Input
- $0.15
- Output
- $0.40
- Cached input
- $0.06
- Input
- $0.16
- Output
- $0.32
- Input
- $0.20
- Output
- $0.40
- Input
- $0.30
- Output
- $0.60
- Input
- $0.361
- Output
- $1.0925
- Cached input
- $0.1805
Source discount 5%; rates as published
- Input
- $0.38
- Output
- $1.15
Endpoint degraded (status -2)
- Input
- $0.75
- Output
- $1.00
- Cached input
- $0.20
- Input
- $0.75
- Output
- $1.00
- Cached input
- $0.25
Model specifications
- Developer
- Modality
- Language
- Context
- 262,144 tokens
- Maximum output
- 16,384 tokens
- Weights
- Open weights
- Provider listed
- Mar 11, 2026
- First discovered
- Oct 16, 2026
Exact identities
- openrouter
google/gemma-4-31b-it - huggingface
google/gemma-4-31B-it - deepinfra
google/gemma-4-31B-it - livebench
gemma-4-31b
Sources
Updated Oct 18, 2026
More from google
View all ↗googleListed 4 days ago
Google: Gemini 4 Argon (batch)
Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep…
1M contextImage inputAudio inputNo reported score
No reported score
googleReleased 18 days ago
Google: Gemini 4 Argon
Long decode can use the unified 1M-token budget for output; reads video, audio and PDFs, with implicit-only caching.
1M contextImage inputAudio input
AA IntelligenceEffort: high · High52.56LiveBenchEffort: high · single run · equal mean of 7 suites84.1%
$1.00in$5.00out/ 1M tokens · USDGoogle AI Studio · via OpenRouter · 8 endpointsgoogle-ai-studio/flex · tier=flex
googleListed 23 days ago
Google: Gemini 3.5 Transcribe
Gemini 3.5 Transcribe is a speech-to-text model from Google. It is suited for synchronous transcription that needs word-level timestamps or speaker diarization, with support for up to eight speakers. Audio...
98.3K contextAudio inputNo reported score
No reported score
googleListed Sep 2
Google: Gemini 3.8 Flash
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
1M contextImage inputAudio inputNo reported score
No reported score
$0.375in$1.875out/ 1M tokens · USDGoogle · via OpenRouter · 9 endpointsgoogle-vertex/global/flex · tier=flex