Model comparison
DeepSeek-V3.1-Terminus vs Granite 4.0 H Small
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 36.5 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeekSummary
- They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Granite 4.0 H Small in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.0 to 43.8.
Side by side
| DeepSeek-V3.1-Terminus | Granite 4.0 H Small | |
|---|---|---|
| Provider | DeepSeek | IBM |
| Noometry Index | 43.1 | 36.5 |
| Released | 2025-09-22 | — |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Granite 4.0 H Small: 36.4 (#209)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Coding | 1426 | 1249 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Granite 4.0 H Small: 24.4 (#163)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1240 |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Granite 4.0 H Small: 31.4 (#223)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Math | 1402 | 1247 |
| Omni-MATH | — | 29.6% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Granite 4.0 H Small: 32.0 (#215)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| MMLU-Pro | — | 56.9% |
| Vectara Hallucination Rate | — | 5.2% |
| GPQA (HELM) | — | 38.3% |
| LMArena Expert | — | 1251 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Granite 4.0 H Small: 38.6 (#228)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Non-English | 1407 | 1216 |
| LMArena Russian | 1436 | 1201 |
| LMArena Chinese | — | 1249 |
| LMArena Spanish | — | 1258 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Granite 4.0 H Small: 68.7 (#183)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Instruction Following | 1404 | 1222 |
| IFEval | — | 89% |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Granite 4.0 H Small: 37.7 (#213)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Longer Query | 1421 | 1242 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Granite 4.0 H Small: 43.8 (#227)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.0 H Small |
|---|---|---|
| LMArena Text | 1419 | 1241 |
| LMArena Creative Writing | 1403 | 1211 |
| LMArena Multi-Turn | 1411 | 1242 |
| WildBench | — | 73.9% |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Granite 4.0 H Small?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 36.5 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Granite 4.0 H Small better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 36.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Granite 4.0 H Small share?
10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Granite 4.0 H Small has 19.