Model comparison
DeepSeek-V3.1-Terminus vs Granite 4.2 3b
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 39.0 on the Noometry Index.
Last verified . 9 shared benchmarks.
DeepSeekSummary
- They share 9 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 6 categories and Granite 4.2 3b in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 47.9.
Side by side
| DeepSeek-V3.1-Terminus | Granite 4.2 3b | |
|---|---|---|
| Provider | DeepSeek | IBM |
| Noometry Index | 42.6 | 39.0 |
| Released | 2025-09-22 | — |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 11 |
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: 41.5 (#114), Granite 4.2 3b: 39.4 (#146)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Coding | 1426 | 1363 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Reasoning Too close to call
DeepSeek-V3.1-Terminus: 25.7 (#126), Granite 4.2 3b: 25.0 (#136)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Hard Prompts | 1427 | 1308 |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
Math Not comparable
DeepSeek-V3.1-Terminus: 35.7 (#138), Granite 4.2 3b: —
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Math | 1402 | — |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Granite 4.2 3b: 35.6 (#169)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Expert | — | 1313 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.3 (#87), Granite 4.2 3b: 42.4 (#193)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Non-English | 1407 | 1268 |
| LMArena Russian | 1434 | 1249 |
| LMArena Chinese | — | 1268 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.4 (#96), Granite 4.2 3b: 67.2 (#195)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Instruction Following | 1406 | 1271 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.5 (#88), Granite 4.2 3b: 39.3 (#179)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Longer Query | 1426 | 1293 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.3 (#87), Granite 4.2 3b: 47.9 (#202)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 4.2 3b |
|---|---|---|
| LMArena Text | 1420 | 1295 |
| LMArena Creative Writing | 1404 | 1215 |
| LMArena Multi-Turn | 1411 | 1293 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Granite 4.2 3b?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 39.0 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Granite 4.2 3b better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 39.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Granite 4.2 3b share?
9 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Granite 4.2 3b has 11.