Model comparison
DeepSeek-V3.1-Terminus vs Inkling-Small
Inkling-Small is the stronger model overall, scoring 46.5 to 43.1 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeekSummary
- They share 12 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 4 categories and Inkling-Small in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 38.6 to 26.4.
- The biggest single-benchmark swing is SciCode: 40.6% for DeepSeek-V3.1-Terminus and 48.7% for Inkling-Small.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 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-V3.1-Terminus | Inkling-Small | |
|---|---|---|
| Provider | DeepSeek | Thinking Machines Lab |
| Noometry Index | 43.1 | 46.5 |
| Released | 2025-09-22 | 2026-07-15 |
| Weights | Open | Open |
| Context window | 164K | 524K |
| Max output | 147K | 1.05M |
| Input $ / M tokens | $0.27 | $0.45 |
| Output $ / M tokens | $1 | $1.20 |
| Results tracked | 16 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Inkling-Small: 43.6 (#85)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| SciCode | 40.6% | 48.7% |
| LMArena Coding | 1426 | 1451 |
| LMArena WebDev | — | 1409 |
| ALE-Bench | 745.17 | — |
Reasoning Inkling-Small leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Inkling-Small: 38.6 (#63)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| CritPt | 1.7% | 8.3% |
| LMArena Hard Prompts | 1426 | 1423 |
| ARC-AGI-2 | — | 40.1% |
| Kagi LLM Benchmark | 57.4% | — |
| ARC-AGI-1 | — | 84% |
| Chess Puzzles | — | 18% |
| Mystery Game Puzzles | — | 6% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 150.15 |
Math Inkling-Small leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Inkling-Small: 45.1 (#77)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| LMArena Math | 1402 | 1459 |
| FrontierMath (Tiers 1-3) | — | 46.3% |
| FrontierMath Tier 4 | — | 17.1% |
| OTIS Mock AIME 2024-2025 | — | 90% |
| ProofBench | — | 6% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Inkling-Small: 48.2 (#77)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| GPQA Diamond | — | 88.5% |
| SimpleQA Verified | — | 19.1% |
| LMArena Expert | — | 1442 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, Inkling-Small: 39.1 (#62)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| LMArena Vision | — | 1235 |
Multilingual Too close to call
DeepSeek-V3.1-Terminus: 52.1 (#92), Inkling-Small: 51.7 (#104)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1407 | 1402 |
| LMArena Russian | 1436 | 1391 |
| LMArena Chinese | — | 1465 |
| LMArena French | — | 1436 |
| LMArena German | — | 1405 |
| LMArena Japanese | — | 1405 |
| LMArena Korean | — | 1363 |
| LMArena Spanish | — | 1428 |
Instruction Following Too close to call
DeepSeek-V3.1-Terminus: 74.0 (#106), Inkling-Small: 73.8 (#114)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1404 | 1399 |
Long Context Too close to call
DeepSeek-V3.1-Terminus: 43.4 (#97), Inkling-Small: 42.7 (#118)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1421 | 1401 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Inkling-Small: 59.6 (#107)
| Benchmark | DeepSeek-V3.1-Terminus | Inkling-Small |
|---|---|---|
| LMArena Text | 1419 | 1414 |
| LMArena Creative Writing | 1403 | 1331 |
| LMArena Multi-Turn | 1411 | 1418 |
| EQ-Bench Creative Writing | — | 1491 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Inkling-Small?
Inkling-Small is the stronger model overall, scoring 46.5 to 43.1 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1-Terminus or Inkling-Small?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Inkling-Small lists at $0.45 and $1.20.
Is DeepSeek-V3.1-Terminus or Inkling-Small better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 42.0 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and Inkling-Small share?
12 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Inkling-Small has 33.