Gemma 4 31B vs Inkling 2 Mini
Inkling 2 Mini leads Gemma 4 31B on coding, 39.5 to 21.9 on the Vibe Coding Index, and Inkling 2 Mini costs 1.2x less per output token ($0.325 vs $0.40 per million). Updated November 22, 2026; 8 benchmarks have a result for both.
Gemma 4 31B or Inkling 2 Mini: which to pick
Pick Inkling 2 Mini for coding: it scores 39.5 on the Vibe Coding Index against 21.9 for Gemma 4 31B.
Gemma 4 31B is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
- Shared results
- 8
- Gemma 4 31B only
- 69
- Inkling 2 Mini only
- 21
- Like-for-like categories
- 5 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
Inkling 2 Mini
Inkling 2 Mini scores 39.5 against 21.9.
2 benchmarks behind each score
Intelligence
AA Intelligence Index
Inkling 2 Mini
Inkling 2 Mini leads, 20.1 against 4.4.
Same board, both measured
General
Vals Index
No call
Only Inkling 2 Mini has a Vals Index result, 46.5%.
Directional only
Long documents
Listed context window
Inkling 2 Mini
Inkling 2 Mini takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Inkling 2 Mini
Gemma 4 31B $0.0003, Inkling 2 Mini $0.0002 per turn.
List prices
At a glance
- Coding index
- Gemma 4 31B 21.9 from 2 benchmarks, Inkling 2 Mini 39.5 from 2 benchmarks.
- Widest category
- General, 23.2 points apart on average over 2 shared benchmarks.
- Widest single gap
- GDPval-AA: Gemma 4 31B 627 against Inkling 2 Mini 1319, 43.2 points apart on a 0 to 100 scale.
- Output price
- Gemma 4 31B $0.40 against Inkling 2 Mini $0.325 per million output tokens, 1.2x apart.
- Context window
- Gemma 4 31B takes 262K tokens, Inkling 2 Mini 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Gemma 4 31B | Bars | Inkling 2 Mini | Ties |
|---|---|---|---|---|
| Coding | 0 | 2 | ||
| Intelligence | 0 | 1 | ||
| Reasoning | 0 | 2 | ||
| Knowledge | 0 | 1 | ||
| General | 0 | 2 | ||
| All shared boards | 0 | 8 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index02
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| LiveBench coding | 20.3% | Not measured | Only Gemma 4 31B measured |
| LMArena WebDev | 1370 | Not measured | Only Gemma 4 31B measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 6.6% | Only Inkling 2 Mini measured |
| Vibe Code Bench (Vals AI) | Not measured | 77.9% | Only Inkling 2 Mini measured |
| ProgramBench (Vals AI) | Not measured | 0.5% | Only Inkling 2 Mini measured |
| IOI (Vals AI) | Not measured | 11.6% | Only Inkling 2 Mini measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 37.9% | Only Inkling 2 Mini measured |
| SkillsBench (Vals AI) | Not measured | 43.2% | Only Inkling 2 Mini measured |
| CyberBench (Vals AI) | Not measured | 74.6% | Only Inkling 2 Mini measured |
| SciCode | 45.5% | Not measured | Only Gemma 4 31B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 36.4% | Not measured | Only Gemma 4 31B measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 0.0% | 5.6% | Inkling 2 Mini |
| FrontierCode 2.0 | 0.0% | 15.1% | Inkling 2 Mini |
| WeirdML | 52.3% | Not measured | Only Gemma 4 31B measured |
| ALE-Bench | 925.5 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: coding | 1459 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: software and IT | 1463 | Not measured | Only Gemma 4 31B measured |
| LMArena WebDev: React | 1362 | Not measured | Only Gemma 4 31B measured |
| LMArena WebDev: HTML | 1380 | Not measured | Only Gemma 4 31B measured |
| SciCode | 8.9% | Not measured | Only Gemma 4 31B measured |
| GameBench 2 | 34.9 | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 31 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| AA Intelligence Index | 4.4 | 20.1 | Inkling 2 Mini |
| Epoch Capabilities Index | 144.1 | Not measured | Only Gemma 4 31B measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index11 boards, none shared
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| LiveBench research | 38.1% | Not measured | Only Gemma 4 31B measured |
| LiveBench browser | 0.0% | Not measured | Only Gemma 4 31B measured |
| LiveBench long agentic | 8.5% | Not measured | Only Gemma 4 31B measured |
| SREBench (Vals AI) | Not measured | 0.0% | Only Inkling 2 Mini measured |
| Tax Agent Bench (Vals AI) | Not measured | 14.1% | Only Inkling 2 Mini measured |
| Legal Research Bench (Vals AI) | Not measured | 17.8% | Only Inkling 2 Mini measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 5.0% | Only Inkling 2 Mini measured |
| BioMysteryBench (Vals AI) | Not measured | 47.8% | Only Inkling 2 Mini measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 0.0% | Only Inkling 2 Mini measured |
| Vals Science Index | Not measured | 23.6% | Only Inkling 2 Mini measured |
| τ²-Bench Telecom | 65.5% | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 33 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| LiveBench reasoning | 68.3% | Not measured | Only Gemma 4 31B measured |
| MysteryMechanism (Vals AI) | Not measured | 23.0% | Only Inkling 2 Mini measured |
| GPQA Diamond (Artificial Analysis run) | 85.7% | Not measured | Only Gemma 4 31B measured |
| Humanity's Last Exam (Artificial Analysis run) | 23.6% | Not measured | Only Gemma 4 31B measured |
| HLE-Diamond (Artificial Analysis run) | 7.2% | 15.2% | Inkling 2 Mini |
| CritPt v2, hard subset (Artificial Analysis run) | 4.0% | 17.0% | Inkling 2 Mini |
| GPQA Diamond (Epoch AI run) | 75.8% | Not measured | Only Gemma 4 31B measured |
| LMArena Text: hard prompts | 1447 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: expert | 1465 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: hard prompts in English | 1456 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: diagrams | 1287 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: homework | 1310 | Not measured | Only Gemma 4 31B measured |
| Kagi LLM Benchmark | 63.5% | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 13 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| OTIS Mock AIME | 73.3% | Not measured | Only Gemma 4 31B measured |
| LMArena Text: math | 1461 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: mathematical work | 1464 | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 11 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| CaseLaw (Vals AI) | 52.6% | Not measured | Only Gemma 4 31B measured |
| MortgageTax (Vals AI) | 61.4% | Not measured | Only Gemma 4 31B measured |
| MedCode (Vals AI) | Not measured | 25.7% | Only Inkling 2 Mini measured |
| AA-Omniscience | -48.5 | -8.6 | Inkling 2 Mini |
| SimpleQA Verified | 10.4% | Not measured | Only Gemma 4 31B measured |
| LMArena Text: science | 1455 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: medicine and healthcare | 1450 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: legal and government | 1438 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: business and finance | 1431 | Not measured | Only Gemma 4 31B measured |
| FACTS Benchmark Suite | 50.1% | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 11 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| Design Arena: UI components | Not measured | 1292 | Only Inkling 2 Mini measured |
| Design Arena: game dev | Not measured | 1287 | Only Inkling 2 Mini measured |
| Design Arena: 3D | Not measured | 1308 | Only Inkling 2 Mini measured |
| LiveBench design | 56.5% | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 5 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| Vals Index | Not measured | 46.5% | Only Inkling 2 Mini measured |
| LiveBench data analysis | 53.3% | Not measured | Only Gemma 4 31B measured |
| SAGE (Vals AI) | 82.8% | 85.9% | Inkling 2 Mini |
| MedScribe (Vals AI) | Not measured | 86.4% | Only Inkling 2 Mini measured |
| IFBench | 75.6% | Not measured | Only Gemma 4 31B measured |
| Long context reasoning (AA-LCR) | 69.7% | Not measured | Only Gemma 4 31B measured |
| GDPval-AA | 627 | 1319 | Inkling 2 Mini |
| DTBench | 82.7% | Not measured | Only Gemma 4 31B measured |
| LMCA | 39.3% | Not measured | Only Gemma 4 31B measured |
| Surface Evolver Bench | 30.6% | Not measured | Only Gemma 4 31B measured |
| LMArena Text | 1442 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: instruction following | 1432 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: creative writing | 1415 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision | 1279 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: English | 1451 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: non-English | 1430 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: Chinese | 1476 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: French | 1435 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: Spanish | 1443 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: Russian | 1460 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: Polish | 1431 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: multi-turn | 1452 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: longer queries | 1445 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: writing and language | 1431 | Not measured | Only Gemma 4 31B measured |
| LMArena Text: entertainment and media | 1404 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: English | 1274 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: Chinese | 1327 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: reading text | 1285 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: humor | 1280 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: captioning | 1210 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: recognizing things | 1228 | Not measured | Only Gemma 4 31B measured |
| LMArena Vision: creative writing | 1287 | Not measured | Only Gemma 4 31B measured |
| LMArena Document | 1434 | Not measured | Only Gemma 4 31B measured |
| EQ-Bench 4 | 1110 | Not measured | Only Gemma 4 31B measured |
| Neither model has a result on the other 15 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemma 4 31B | Inkling 2 Mini | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Inkling 2 Mini costs less
Repository review
50K input + 3K output tokens
Inkling 2 Mini costs less
Agent loop
200K input + 10K output tokens
Inkling 2 Mini costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Google Gemma 4 31B | Thinking Machines Inkling 2 Mini |
|---|---|---|
| Maker | Thinking Machines | |
| Weights | Open weights | Open weights |
| Released | April 2, 2026 | November 21, 2026 |
| Context window | 262K | 1.05M |
| Input $/Mtok | $0.14 | $0.0775 |
| Output $/Mtok | $0.40 | $0.325 |
| Coding index | 21.9 | 39.5 |
| Independent measurements behind it | 2 | 2 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemma 4 31B better than Inkling 2 Mini for coding?
Pick Inkling 2 Mini for coding: it scores 39.5 on the Vibe Coding Index against 21.9 for Gemma 4 31B. Gemma 4 31B is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
Which is cheaper, Gemma 4 31B or Inkling 2 Mini?
Gemma 4 31B costs $0.14 per million input tokens and $0.40 per million output tokens; Inkling 2 Mini costs $0.0775 and $0.325. Cost per solved task is known for neither of them.
Which has the bigger context window, Gemma 4 31B or Inkling 2 Mini?
Inkling 2 Mini, with 1.05M tokens against 262K.
How many benchmarks is the Gemma 4 31B vs Inkling 2 Mini comparison based on?
8 benchmarks have a result for both models as of November 22, 2026. Gemma 4 31B's coding index rests on 2 independent benchmarks and Inkling 2 Mini's on 2. Where only one model has a result, the page says so instead of naming a winner.