Gemma 4 31B benchmarks
Gemma 4 31B is an open-weights model from Google, released 2 Apr 2026. It ranks 50th of 51 on our overall leaderboard and 35th of 50 for value. At A$0.29 per million tokens (blended), it costs about 13× cheaper than the median model we track. It writes about 35 tokens a second, and answers start after 50 s on average, thinking included. Its weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region.
- CONTEXT WINDOW
- 256,000 tokens
- INPUT
- text, image
- WEIGHTS
- Open (Apache 2.0)
- RELEASED
- 2 Apr 2026
- OUTPUT SPEED
- 35 tok/s
- ANSWER STARTS AFTER
- 50 s (thinking included)
Gemma 4 31B across our rankings.
Benchmark results.
Independent test scores are for the reasoning setting. “Among tracked” ranks Gemma 4 31B against the 51 models on this site; LMArena ranks run across its full leaderboard.
| BENCHMARK | RESULT | AMONG TRACKED | SOURCE DETAIL |
|---|---|---|---|
| INDEPENDENT TESTS · ARTIFICIAL ANALYSIS | |||
| AA Intelligence Index | 19.0 | #50 of 51 | Reasoning |
| AA Coding Index | 43.4 | #42 of 42 | Reasoning |
| GPQA Diamond | 85.7% | #42 of 44 | Reasoning |
| Humanity’s Last Exam | 23.6% | #49 of 51 | Reasoning |
| AA-LCR | 69.7% | #50 of 51 | Reasoning |
| Terminal-Bench 2.1 | 43.4% | #42 of 42 | Reasoning |
| τ²-Bench (banking) | 14.8% | #39 of 41 | Reasoning |
| SciCode | 45.5% | #44 of 46 | Reasoning |
| BLIND HUMAN VOTES · LMARENA | |||
| LMArena Text5,894 votes · ±8 | 1451 | #38 of 42 | #68 of 402 on LMArena · 13 Sep 2026 |
| LMArena WebDev11,584 votes · ±7 | 1364 | #44 of 45 | #91 of 129 on LMArena · 22 Sep 2026 |
| LMArena Vision35,713 votes · ±6 | 1261 | #28 of 33 | #38 of 152 on LMArena · 13 Sep 2026 |
| LMArena Document12,326 votes · ±8 | 1445 | #25 of 26 | #34 of 44 on LMArena · 13 Sep 2026 |
What Gemma 4 31B costs in Australian dollars.
List prices converted at the snapshot’s RBA rate, excluding GST. Reasoning models are billed for their thinking as output tokens, so heavier settings cost more than the job estimates show.
| PER MILLION TOKENS | AUD | USD LIST |
|---|---|---|
| Input tokens | A$0.20 | US$0.14 |
| Output tokens | A$0.56 | US$0.40 |
| Blended (3 in : 1 out) | A$0.29 | US$0.20 |
| JOB | COST (AUD) | MEDIAN MODEL |
|---|---|---|
| Customer support replyPER 1,000 REPLIES | A$0.56 | ≈ A$6.58 |
| Summarise a 30-page documentPER 100 DOCUMENTS | A$0.44 | ≈ A$4.74 |
| Agentic coding taskPER 10 TASKS | A$0.34 | ≈ A$3.72 |
| SETTING | INTELLIGENCE | A$ / 1M | SPEED |
|---|---|---|---|
| Reasoning | 19.0 | 35 tok/s | |
| Non-reasoning | 13.9 | A$0.29 | 38 tok/s |
Running Gemma 4 31B in Australia.
Its weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region. How the platforms compare →
| PLATFORM | AVAILABILITY | DETAIL |
|---|---|---|
| AWS Bedrock · Sydney | NOT IN AU | |
| AWS Bedrock · Melbourne | NOT IN AU | |
| Azure · Australia East | NOT OFFERED | |
| Google Vertex AI · Sydney | NOT OFFERED | |
| Your own infrastructure | SELF-HOST | Open weights: run it on your own servers or GPU instances in an Australian region. |
Compare it with.
Gemma 4 31B, answered.
How much does Gemma 4 31B cost in Australian dollars?
Gemma 4 31B lists at US$0.14 per million input tokens and US$0.40 per million output tokens — A$0.20 and A$0.56 at A$1 = US$0.7123 (RBA, 22 Sep 2026), excluding GST. A typical customer support reply works out at about A$0.56 per 1,000 replies, before any reasoning tokens.
Can I use Gemma 4 31B in Australia with data kept onshore?
Gemma 4 31B’s weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region. Availability comes from the cloud providers’ own documentation; check your provider’s current terms before relying on it for data-residency obligations.
Is Gemma 4 31B good for coding?
It ranks 41st of 42 on our coding ranking, with an Artificial Analysis Coding Index of 43.4 and an LMArena WebDev rating of 1364 (91st of 129). The current leader is Claude Fable 5.1.
How good is Gemma 4 31B at agent and automation work?
It ranks 41st of 41 on our agents ranking, completing 14.8% of τ²-Bench banking customer-service tasks and 43.4% of Terminal-Bench 2.1 tasks.
How fast is Gemma 4 31B?
Artificial Analysis measures Gemma 4 31B at about 35 tok/s of output, with the answer starting after 50 s on average once thinking time is included (reasoning setting).
What is Gemma 4 31B’s context window?
Google documents a context window of 256,000 tokens (250K). On AA-LCR, which tests reasoning across ~100,000-token document sets, it scores 69.7%.
Ratings: LMArena leaderboard dataset (CC BY 4.0), rescaled for the lens scores · leaderboards to 22 Sep 2026
Evaluations, prices and speed: Artificial Analysis (artificialanalysis.ai) · fetched 23 Sep 2026
Exchange rate: Reserve Bank of Australia, table F11.1 · A$1 = US$0.7123 on 22 Sep 2026 · prices exclude GST
Snapshot 23 Sep 2026 · updated weekly · How the rankings work →