Model comparison
DeepSeek-R1 vs Inkling-Small
Inkling-Small is the stronger model overall, scoring 46.5 to 42.3 on the Noometry Index.
Last verified . 25 shared benchmarks.
DeepSeekSummary
- They share 25 benchmarks with published results for both. DeepSeek-R1 scores higher in 4 categories and Inkling-Small in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 38.6 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 84% for Inkling-Small.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $0.50 / $2.15 for DeepSeek-R1.
- Inkling-Small accepts more context: 524K tokens versus 164K.
- Inkling-Small has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1 | Inkling-Small | |
|---|---|---|
| Provider | DeepSeek | Thinking Machines Lab |
| Noometry Index | 42.3 | 46.5 |
| Released | 2025-01-20 | 2026-07-15 |
| Weights | Proprietary | Open |
| Context window | 164K | 524K |
| Max output | 64K | 1.05M |
| Input $ / M tokens | $0.50 | $0.45 |
| Output $ / M tokens | $2.15 | $1.20 |
| Results tracked | 52 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), Inkling-Small: 43.6 (#85)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| SciCode | 35.7% | 48.7% |
| LMArena Coding | 1427 | 1451 |
| Aider Polyglot | 71.4% | — |
| LMArena WebDev | — | 1409 |
| WeirdML | 41.6% | — |
| LiveBench Coding | 66.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use Not comparable
DeepSeek-R1: 30.7 (#75), Inkling-Small: —
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
Reasoning Inkling-Small leads
DeepSeek-R1: 18.6 (#278), Inkling-Small: 38.6 (#63)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| ARC-AGI-2 | 1.3% | 40.1% |
| ARC-AGI-1 | 21.2% | 84% |
| CritPt | 1.1% | 8.3% |
| LMArena Hard Prompts | 1416 | 1423 |
| Epoch Capabilities Index | 141.29 | 150.15 |
| SimpleBench | 40.8% | — |
| Kagi LLM Benchmark | 69.4% | — |
| Chess Puzzles | — | 18% |
| LiveBench Reasoning | 83.2% | — |
| Mystery Game Puzzles | — | 6% |
| LiveBench Data Analysis | 69.8% | — |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math Inkling-Small leads
DeepSeek-R1: 43.8 (#79), Inkling-Small: 45.1 (#77)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.4% | 90% |
| LMArena Math | 1400 | 1459 |
| FrontierMath (Tiers 1-3) | — | 46.3% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 6% |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge Inkling-Small leads
DeepSeek-R1: 44.5 (#87), Inkling-Small: 48.2 (#77)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| GPQA Diamond | 76.3% | 88.5% |
| LMArena Expert | 1394 | 1442 |
| SimpleQA Verified | — | 19.1% |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multimodal Not comparable
DeepSeek-R1: —, Inkling-Small: 39.1 (#62)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| LMArena Vision | — | 1235 |
Multilingual Too close to call
DeepSeek-R1: 52.4 (#85), Inkling-Small: 51.7 (#104)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1412 | 1402 |
| LMArena Chinese | 1442 | 1465 |
| LMArena French | 1417 | 1436 |
| LMArena German | 1404 | 1405 |
| LMArena Japanese | 1391 | 1405 |
| LMArena Korean | 1360 | 1363 |
| LMArena Russian | 1423 | 1391 |
| LMArena Spanish | 1411 | 1428 |
Instruction Following Inkling-Small leads
DeepSeek-R1: 72.0 (#143), Inkling-Small: 73.8 (#114)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1382 | 1399 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), Inkling-Small: 42.7 (#118)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1391 | 1401 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), Inkling-Small: 59.6 (#107)
| Benchmark | DeepSeek-R1 | Inkling-Small |
|---|---|---|
| LMArena Text | 1428 | 1414 |
| LMArena Creative Writing | 1405 | 1331 |
| EQ-Bench Creative Writing | 1500 | 1491 |
| LMArena Multi-Turn | 1405 | 1418 |
| Short-Story Creative Writing | 83% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than Inkling-Small?
Inkling-Small is the stronger model overall, scoring 46.5 to 42.3 on the Noometry Index.
Which is cheaper, DeepSeek-R1 or Inkling-Small?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; DeepSeek-R1 lists at $0.50 and $2.15.
Is DeepSeek-R1 or Inkling-Small better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 43.6 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 164K.
How many benchmarks do DeepSeek-R1 and Inkling-Small share?
25 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Inkling-Small has 33.