# Gemma 4 31B vs Inkling 2 Mini

> Side-by-side benchmark comparison for Gemma 4 31B and Inkling 2 Mini across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gemma-4-31b-vs-inkling-2-mini
- Last updated: November 21, 2026

- Shared sourced benchmarks: 3
- HTML indexing: indexable
- Ranking lane: BenchAlign v6.0

## Quick Verdict

Evidence is too thin for a benchmark verdict: Gemma 4 31B and Inkling 2 Mini share only 3 sourced benchmark rows.

## Summary

- Only 3 shared sourced benchmark rows are available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Gemma 4 31B has the lower published output-token price.
- Inkling 2 Mini has the larger published context window at 1M.

## Model Snapshot

| Property | Gemma 4 31B | Inkling 2 Mini |
|----------|----------|----------|
| Creator | Google | Thinking Machines Lab |
| Type | Open Weight | Open Weight |
| Reasoning | Reasoning | Hybrid |
| Context | 256K | 1M |
| Independent public score (not head-to-head) | 43 | 55.9 |
| Benchmarks Covered | 13 | 33 |
| Pricing (input/output) | $0.00 / $0.00 | $0.15 / $0.50 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 33.5 (Supported · #154/179)
- Inkling 2 Mini public-lane score: 59.1 (Estimated · #61/179)

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| b3 Agent Security | 83.0% | Coming soon | Coming soon |
| Cyber Misuse Refusal | 84.0% | Coming soon | Coming soon |
| Gert Labs | 35.26% | Coming soon | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 5.60% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 71.6% | Coming soon |
| Terminal-Bench-Science 0.1 | Coming soon | 1.4% | Coming soon |
| BrowseComp | Coming soon | 79% | Coming soon |
| DeepSearchQA | Coming soon | 85.3% | Coming soon |
| MCP Atlas | Coming soon | 82.7% | Coming soon |
| JobBench | Coming soon | 49.6% | Coming soon |
| EnterpriseOps-Gym | Coming soon | 42.00% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 43.5 (Supported · #114/179)
- Inkling 2 Mini public-lane score: 53.9 (Estimated · #67/179)

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench v6 | 80.0% | Coming soon | Coming soon |
| SWE-Rebench | 41.6% | Coming soon | Coming soon |
| React Native Evals | 75.2% | Coming soon | Coming soon |
| FrontierCode 2.0 | Coming soon | 15.1% | Coming soon |
| DeepSWE | Coming soon | 48.2% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 71.6% | Coming soon |
| SWE Multilingual | Coming soon | 73.7% | Coming soon |
| SWE-bench Verified | Coming soon | 79.2% | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 41.8 (Supported · #111/161)
- Inkling 2 Mini public-lane score: 57.8 (Estimated · #58/161)

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| MRCRv2 | 66.4% | Coming soon | Coming soon |
| LongBench v2 | Coming soon | 66.9% | Coming soon |
| MRCR v2 256K-512K | Coming soon | 46.4% | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 44.4 (Supported · #89/119)
- Inkling 2 Mini public-lane score: 55.5 (Estimated · #62/119)

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| ParseBench-1 | 62.40% | Coming soon | Coming soon |
| MMMU-Pro | 76.9% | 74.9% | Gemma 4 31B |
| PerceptionBench | Coming soon | 57.5% | Coming soon |
| OmniDocBench 1.6 | Coming soon | 92.8% | Coming soon |
| BabyVision | Coming soon | 53.4% | Coming soon |
| BabyVision w/ Python | Coming soon | 88.8% | Coming soon |
| ScreenSpot Pro | Coming soon | 76.5% | Coming soon |
| CharXiv | Coming soon | 88.3% | Coming soon |
| CharXiv w/o tools | Coming soon | 80.6% | Coming soon |
| Chartography (no tools) | Coming soon | 22.1% | Coming soon |
| Chartography (tools) | Coming soon | 72.5% | Coming soon |
| ZeroBench | Coming soon | 17.0% | Coming soon |
| ZeroBench w/ Python | Coming soon | 34.0% | Coming soon |
| MathVision | Coming soon | 89.0% | Coming soon |

### Knowledge

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 43.6 (Supported · #111/158)
- Inkling 2 Mini public-lane score: 51.9 (Estimated · #79/158)

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 84.3% | 86.6% | Inkling 2 Mini |
| MMLU-Pro | 85.2% | Coming soon | Coming soon |
| HLE | 26.5% | Coming soon | Coming soon |
| HLE w/o tools | 19.5% | 30.7% | Inkling 2 Mini |
| GPQA-D | Coming soon | 86.6% | Coming soon |

### Multilingual

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 41.3 (Supported · #61/87)
- Inkling 2 Mini public-lane score: 54.7 (Estimated · #39/87)

### Instruction Following

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 41.6 (Supported · #102/143)
- Inkling 2 Mini public-lane score: Coming soon

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| MultiChallenge (provider reports) | Coming soon | 50.00% | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- Gemma 4 31B public-lane score: 40.9 (Estimated · #94/125)
- Inkling 2 Mini public-lane score: Coming soon

| Benchmark | Gemma 4 31B | Inkling 2 Mini | Winner |
|-----------|-----------|-----------|--------|
| Apex | Coming soon | 34.4% | Coming soon |

## FAQ

### Is there enough evidence to pick between Gemma 4 31B and Inkling 2 Mini?

Not yet. The models share only 3 sourced benchmark rows, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [Gemma 4 31B vs Gemma 4 26B A4B](/compare/gemma-4-26b-a4b-vs-gemma-4-31b)
- [Inkling 2 Mini vs Gemma 4 26B A4B](/compare/gemma-4-26b-a4b-vs-inkling-2-mini)
- [Gemma 4 31B vs Gemma 4 E2B](/compare/gemma-4-31b-vs-gemma-4-e2b)
- [Inkling 2 Mini vs Gemma 4 E2B](/compare/gemma-4-e2b-vs-inkling-2-mini)

## Explore More

- [Gemma 4 31B profile](/models/gemma-4-31b)
- [Inkling 2 Mini profile](/models/inkling-2-mini)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)
