Model comparison
DBRX vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 29.4 on the Noometry Index.
Last verified . 17 shared benchmarks.
Databricks
StepFunSummary
- They share 17 benchmarks with published results for both. DBRX scores higher in 0 categories and Step 3.5 Flash in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3.5 Flash leads 58.8 to 33.6.
Side by side
| DBRX | Step 3.5 Flash | |
|---|---|---|
| Provider | Databricks | StepFun |
| Noometry Index | 29.4 | 42.3 |
| Released | 2024-03-27 | 2026-01-29 |
| Weights | Open | Open |
| Context window | — | 256K |
| Max output | — | 256K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.30 |
| Results tracked | 21 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
DBRX: 32.9 (#266), Step 3.5 Flash: 42.4 (#105)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1132 | 1436 |
| HumanEval+ | 70.1% | — |
| MBPP+ | 55.8% | — |
Reasoning Too close to call
DBRX: 21.4 (#222), Step 3.5 Flash: 22.2 (#202)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1411 |
| NYT Connections (extended) | — | 28.4% |
Math Step 3.5 Flash leads
DBRX: 24.3 (#269), Step 3.5 Flash: 42.6 (#84)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1145 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| MATH Level 5 | 11.7% | — |
Knowledge Step 3.5 Flash leads
DBRX: 14.9 (#294), Step 3.5 Flash: 39.6 (#132)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1076 | 1421 |
| GPQA Diamond | 32.9% | — |
Multilingual Step 3.5 Flash leads
DBRX: 29.3 (#268), Step 3.5 Flash: 50.5 (#119)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1071 | 1385 |
| LMArena Chinese | 1068 | 1447 |
| LMArena French | 1096 | 1421 |
| LMArena German | 1057 | 1405 |
| LMArena Japanese | 990 | 1354 |
| LMArena Korean | 993 | 1352 |
| LMArena Russian | 1078 | 1385 |
| LMArena Spanish | 1064 | 1419 |
Instruction Following Step 3.5 Flash leads
DBRX: 57.5 (#270), Step 3.5 Flash: 73.1 (#124)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1112 | 1385 |
Long Context Step 3.5 Flash leads
DBRX: 33.7 (#258), Step 3.5 Flash: 42.8 (#117)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1112 | 1402 |
Writing & Preference Step 3.5 Flash leads
DBRX: 33.6 (#275), Step 3.5 Flash: 58.8 (#113)
| Benchmark | DBRX | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1119 | 1403 |
| LMArena Creative Writing | 1104 | 1357 |
| LMArena Multi-Turn | 1111 | 1405 |
Frequently asked questions
Is DBRX better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 29.4 on the Noometry Index.
Is DBRX or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 32.9 in the Noometry coding category.
How many benchmarks do DBRX and Step 3.5 Flash share?
17 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and Step 3.5 Flash has 19.