Model comparison
DeepSeek-V2.5 (Sep 2024) vs Step 2 16k Exp 202412
Step 2 16k Exp 202412 is the stronger model overall, scoring 39.2 to 37.6 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeek
StepFunSummary
- They share 12 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 0 categories and Step 2 16k Exp 202412 in 8 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 2 16k Exp 202412 leads 38.5 to 31.7.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 37.6 | 39.2 |
| Released | 2024-09-06 | — |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 2 16k Exp 202412 leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Step 2 16k Exp 202412: 38.5 (#174)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Coding | 1309 | 1316 |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning Too close to call
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Step 2 16k Exp 202412: 25.9 (#142)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1299 |
Math Too close to call
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Step 2 16k Exp 202412: 36.3 (#169)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Math | 1288 | 1304 |
Knowledge Too close to call
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Step 2 16k Exp 202412: 35.2 (#187)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Expert | 1266 | 1279 |
Multilingual Step 2 16k Exp 202412 leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Step 2 16k Exp 202412: 43.7 (#181)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Non-English | 1273 | 1290 |
| LMArena Chinese | 1318 | 1331 |
| LMArena Russian | 1289 | 1324 |
| LMArena French | 1289 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Spanish | 1248 | — |
Instruction Following Too close to call
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Step 2 16k Exp 202412: 67.9 (#192)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Instruction Following | 1280 | 1287 |
Long Context Too close to call
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Step 2 16k Exp 202412: 39.7 (#170)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Longer Query | 1301 | 1307 |
Writing & Preference Step 2 16k Exp 202412 leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Step 2 16k Exp 202412: 52.3 (#173)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Text | 1294 | 1321 |
| LMArena Creative Writing | 1285 | 1328 |
| LMArena Multi-Turn | 1297 | 1292 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Step 2 16k Exp 202412?
Step 2 16k Exp 202412 is the stronger model overall, scoring 39.2 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Step 2 16k Exp 202412 better for coding?
Step 2 16k Exp 202412 scores higher on coding benchmarks: 38.5 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Step 2 16k Exp 202412 share?
12 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Step 2 16k Exp 202412 has 12.