Model comparison
DeepSeek-V3.1-Terminus vs Step 2 16k Exp 202412
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 39.2 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeek
StepFunSummary
- They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Step 2 16k Exp 202412 in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.0 to 52.3.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 43.1 | 39.2 |
| Released | 2025-09-22 | — |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 12 |
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), Step 2 16k Exp 202412: 38.5 (#174)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Coding | 1426 | 1316 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Reasoning Too close to call
DeepSeek-V3.1-Terminus: 26.4 (#133), Step 2 16k Exp 202412: 25.9 (#142)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1299 |
| 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), Step 2 16k Exp 202412: 36.3 (#169)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Math | 1402 | 1304 |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Step 2 16k Exp 202412: 35.2 (#187)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Expert | — | 1279 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Step 2 16k Exp 202412: 43.7 (#181)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Non-English | 1407 | 1290 |
| LMArena Russian | 1436 | 1324 |
| LMArena Chinese | — | 1331 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Step 2 16k Exp 202412: 67.9 (#192)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1287 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Step 2 16k Exp 202412: 39.7 (#170)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Longer Query | 1421 | 1307 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Step 2 16k Exp 202412: 52.3 (#173)
| Benchmark | DeepSeek-V3.1-Terminus | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Text | 1419 | 1321 |
| LMArena Creative Writing | 1403 | 1328 |
| LMArena Multi-Turn | 1411 | 1292 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Step 2 16k Exp 202412?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 39.2 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Step 2 16k Exp 202412 better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 38.5 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Step 2 16k Exp 202412 share?
10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Step 2 16k Exp 202412 has 12.