Model comparison
DeepSeek-V3 vs Inkling-Small
Inkling-Small is the stronger model overall, scoring 46.5 to 39.5 on the Noometry Index. DeepSeek-V3 costs 1.6× less per token, which makes it the better buy when Inkling-Small's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
DeepSeekSummary
- They share 23 benchmarks with published results for both. DeepSeek-V3 scores higher in 0 categories and Inkling-Small in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 38.6 to 20.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for DeepSeek-V3 and 90% for Inkling-Small.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $0.45 / $1.20 for Inkling-Small.
- Inkling-Small accepts more context: 524K tokens versus 164K.
Side by side
| DeepSeek-V3 | Inkling-Small | |
|---|---|---|
| Provider | DeepSeek | Thinking Machines Lab |
| Noometry Index | 39.5 | 46.5 |
| Released | 2024-12-26 | 2026-07-15 |
| Weights | Open | Open |
| Context window | 164K | 524K |
| Max output | 164K | 1.05M |
| Input $ / M tokens | $0.24 | $0.45 |
| Output $ / M tokens | $0.90 | $1.20 |
| Results tracked | 60 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
DeepSeek-V3: 42.3 (#106), Inkling-Small: 43.6 (#85)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| SciCode | 35.8% | 48.7% |
| LMArena Coding | 1368 | 1451 |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1409 |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Inkling-Small: —
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| METR Time Horizons | 49.6% | — |
Reasoning Inkling-Small leads
DeepSeek-V3: 20.5 (#236), Inkling-Small: 38.6 (#63)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| CritPt | 0% | 8.3% |
| LMArena Hard Prompts | 1365 | 1423 |
| Epoch Capabilities Index | 135.94 | 150.15 |
| ARC-AGI-2 | — | 40.1% |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| ARC-AGI-1 | — | 84% |
| Chess Puzzles | — | 18% |
| LiveBench Reasoning | 65.8% | — |
| Mystery Game Puzzles | — | 6% |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| BIG-Bench Hard | 87.5% | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |
Math Inkling-Small leads
DeepSeek-V3: 32.1 (#219), Inkling-Small: 45.1 (#77)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 90% |
| LMArena Math | 1373 | 1459 |
| FrontierMath (Tiers 1-3) | — | 46.3% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 6% |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Inkling-Small leads
DeepSeek-V3: 37.5 (#155), Inkling-Small: 48.2 (#77)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| GPQA Diamond | 67.6% | 88.5% |
| LMArena Expert | 1351 | 1442 |
| SimpleQA Verified | — | 19.1% |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |
Multimodal Not comparable
DeepSeek-V3: —, Inkling-Small: 39.1 (#62)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| LMArena Vision | — | 1235 |
Multilingual Inkling-Small leads
DeepSeek-V3: 48.5 (#143), Inkling-Small: 51.7 (#104)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1358 | 1402 |
| LMArena Chinese | 1391 | 1465 |
| LMArena French | 1385 | 1436 |
| LMArena German | 1374 | 1405 |
| LMArena Japanese | 1333 | 1405 |
| LMArena Korean | 1319 | 1363 |
| LMArena Russian | 1373 | 1391 |
| LMArena Spanish | 1358 | 1428 |
Instruction Following Inkling-Small leads
DeepSeek-V3: 72.8 (#130), Inkling-Small: 73.8 (#114)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1345 | 1399 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context Inkling-Small leads
DeepSeek-V3: 34.0 (#253), Inkling-Small: 42.7 (#118)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1352 | 1401 |
| Fiction.LiveBench | 50% | — |
Writing & Preference Inkling-Small leads
DeepSeek-V3: 57.4 (#130), Inkling-Small: 59.6 (#107)
| Benchmark | DeepSeek-V3 | Inkling-Small |
|---|---|---|
| LMArena Text | 1375 | 1414 |
| LMArena Creative Writing | 1364 | 1331 |
| EQ-Bench Creative Writing | 1472 | 1491 |
| LMArena Multi-Turn | 1389 | 1418 |
| Short-Story Creative Writing | 77% | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |
Frequently asked questions
Is DeepSeek-V3 better than Inkling-Small?
Inkling-Small is the stronger model overall, scoring 46.5 to 39.5 on the Noometry Index. DeepSeek-V3 costs 1.6× less per token, which makes it the better buy when Inkling-Small's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3 or Inkling-Small?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Inkling-Small lists at $0.45 and $1.20.
Is DeepSeek-V3 or Inkling-Small better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 164K.
How many benchmarks do DeepSeek-V3 and Inkling-Small share?
23 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Inkling-Small has 33.