Model comparison
DeepSeek-V3.1-Terminus vs Granite 3.1 8b Instruct
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 32.4 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 3.1 8b Instruct 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 35.5.
Side by side
| DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct | |
|---|---|---|
| Provider | DeepSeek | IBM |
| Noometry Index | 43.1 | 32.4 |
| 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 | 13 |
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 3.1 8b Instruct: 34.5 (#233)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Coding | 1426 | 1186 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Granite 3.1 8b Instruct: 24.1 (#120)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Granite 3.1 8b Instruct: 22.1 (#207)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1145 |
| 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 3.1 8b Instruct: 33.0 (#209)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Math | 1402 | 1152 |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Granite 3.1 8b Instruct: 31.1 (#220)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Expert | — | 1142 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Granite 3.1 8b Instruct: 30.9 (#260)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Non-English | 1407 | 1099 |
| LMArena Russian | 1436 | 1092 |
| LMArena Chinese | — | 1145 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Granite 3.1 8b Instruct: 58.6 (#259)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Instruction Following | 1404 | 1131 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Granite 3.1 8b Instruct: 35.2 (#241)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Longer Query | 1421 | 1162 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Granite 3.1 8b Instruct: 35.5 (#266)
| Benchmark | DeepSeek-V3.1-Terminus | Granite 3.1 8b Instruct |
|---|---|---|
| LMArena Text | 1419 | 1150 |
| LMArena Creative Writing | 1403 | 1129 |
| LMArena Multi-Turn | 1411 | 1108 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Granite 3.1 8b Instruct?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 32.4 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Granite 3.1 8b Instruct better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 34.5 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Granite 3.1 8b Instruct share?
10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Granite 3.1 8b Instruct has 13.