Model comparison
DeepSeek LLM 67B vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 24.9 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeek
StepFunSummary
- They share 10 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Step 3 leads 36.8 to 7.0.
Side by side
| DeepSeek LLM 67B | Step 3 | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 24.9 | 40.5 |
| Released | 2023-11-29 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
DeepSeek LLM 67B: 31.9 (#278), Step 3: 40.1 (#147)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Coding | 1096 | 1367 |
Reasoning Step 3 leads
DeepSeek LLM 67B: 16.5 (#304), Step 3: 28.4 (#105)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| Chess Puzzles | 0% | — |
| Epoch Capabilities Index | 110.5 | — |
Math Step 3 leads
DeepSeek LLM 67B: 8.7 (#324), Step 3: 37.6 (#148)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Math | 1108 | 1366 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| MATH Level 5 | 6.4% | — |
Knowledge Step 3 leads
DeepSeek LLM 67B: 7.0 (#313), Step 3: 36.8 (#164)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| LMArena Expert | — | 1333 |
Multimodal Not comparable
DeepSeek LLM 67B: —, Step 3: 35.5 (#86)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Step 3 leads
DeepSeek LLM 67B: 29.4 (#267), Step 3: 46.3 (#159)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Non-English | 1073 | 1327 |
| LMArena Chinese | 1132 | 1397 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1269 |
| LMArena Russian | — | 1331 |
| LMArena Spanish | — | 1371 |
Instruction Following Step 3 leads
DeepSeek LLM 67B: 55.4 (#277), Step 3: 70.4 (#164)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1079 | 1332 |
Long Context Step 3 leads
DeepSeek LLM 67B: 33.1 (#265), Step 3: 40.3 (#157)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1092 | 1326 |
Writing & Preference Step 3 leads
DeepSeek LLM 67B: 31.6 (#282), Step 3: 54.3 (#151)
| Benchmark | DeepSeek LLM 67B | Step 3 |
|---|---|---|
| LMArena Text | 1105 | 1350 |
| LMArena Creative Writing | 1067 | 1321 |
| LMArena Multi-Turn | 1082 | 1341 |
Frequently asked questions
Is DeepSeek LLM 67B better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Step 3 share?
10 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Step 3 has 17.