Model comparison
Grok 4.3 vs Step 3.5 Flash
Grok 4.3 is the stronger model overall, scoring 43.6 to 41.7 on the Noometry Index. Step 3.5 Flash costs 10× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
StepFunSummary
- They share 18 benchmarks with published results for both. Grok 4.3 scores higher in 3 categories and Step 3.5 Flash in 5 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.3 leads 34.9 to 21.2.
- The biggest single-benchmark swing is NYT Connections (extended): 55.2% for Grok 4.3 and 28.4% for Step 3.5 Flash.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 256K.
- Step 3.5 Flash has downloadable open weights; the other is API-only.
Side by side
| Grok 4.3 | Step 3.5 Flash | |
|---|---|---|
| Provider | xAI | StepFun |
| Noometry Index | 43.6 | 41.7 |
| Released | 2026-04-17 | 2026-01-29 |
| Weights | Proprietary | Open |
| Context window | 1M | 256K |
| Max output | 30K | 256K |
| Input $ / M tokens | $1.25 | $0.10 |
| Output $ / M tokens | $2.50 | $0.30 |
| Results tracked | 40 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Grok 4.3: 41.3 (#117), Step 3.5 Flash: 41.7 (#106)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1415 | 1436 |
| LMArena WebDev | 1357 | — |
| SciCode | 47.3% | — |
| WeirdML | 49.9% | — |
| ALE-Bench | 944.17 | — |
Agentic & Tool Use Not comparable
Grok 4.3: 27.6 (#93), Step 3.5 Flash: —
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| GDP.pdf | 8% | — |
| LMArena Search | 1165 | — |
| Vending-Bench 2 | 35.26 | — |
Reasoning Grok 4.3 leads
Grok 4.3: 34.9 (#60), Step 3.5 Flash: 21.2 (#202)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| NYT Connections (extended) | 55.2% | 28.4% |
| LMArena Hard Prompts | 1397 | 1411 |
| CritPt | 8% | — |
| Chess Puzzles | 25% | — |
| DTBench | 90.7% | — |
| LMCA | 38.3% | — |
| Epoch Capabilities Index | 149.13 | — |
| ForecastBench | 60.3 | — |
Math Grok 4.3 leads
Grok 4.3: 45.8 (#60), Step 3.5 Flash: 40.6 (#77)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1388 | 1408 |
| FrontierMath (Tiers 1-3) | 42.8% | — |
| FrontierMath Tier 4 | 14.6% | — |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| ProofBench | 11% | — |
Knowledge Grok 4.3 leads
Grok 4.3: 52.5 (#57), Step 3.5 Flash: 39.0 (#127)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1385 | 1422 |
| GPQA Diamond | 88.8% | — |
| SimpleQA Verified | 33.2% | — |
Multimodal Not comparable
Grok 4.3: 31.6 (#96), Step 3.5 Flash: —
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Vision | 1231 | — |
| Blueprint-Bench 2 | 0% | — |
Multilingual Too close to call
Grok 4.3: 50.7 (#114), Step 3.5 Flash: 50.7 (#115)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1385 | 1386 |
| LMArena Chinese | 1423 | 1448 |
| LMArena French | 1412 | 1421 |
| LMArena German | 1395 | 1404 |
| LMArena Japanese | 1379 | 1353 |
| LMArena Korean | 1357 | 1350 |
| LMArena Russian | 1399 | 1385 |
| LMArena Spanish | 1402 | 1419 |
Instruction Following Too close to call
Grok 4.3: 72.3 (#135), Step 3.5 Flash: 73.3 (#119)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1366 | 1385 |
Long Context Too close to call
Grok 4.3: 42.5 (#118), Step 3.5 Flash: 42.8 (#112)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1393 | 1402 |
Writing & Preference Too close to call
Grok 4.3: 58.6 (#112), Step 3.5 Flash: 59.0 (#108)
| Benchmark | Grok 4.3 | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1397 | 1403 |
| LMArena Creative Writing | 1380 | 1357 |
| LMArena Multi-Turn | 1406 | 1405 |
| EQ-Bench 4 | 1075 | — |
Frequently asked questions
Is Grok 4.3 better than Step 3.5 Flash?
Grok 4.3 is the stronger model overall, scoring 43.6 to 41.7 on the Noometry Index. Step 3.5 Flash costs 10× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Which is cheaper, Grok 4.3 or Step 3.5 Flash?
Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.
Is Grok 4.3 or Step 3.5 Flash better for coding?
They score almost the same on coding (41.3 vs 41.7); test both on your own repository before choosing.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 256K.
How many benchmarks do Grok 4.3 and Step 3.5 Flash share?
18 benchmarks have published results for both models. Grok 4.3 has 40 scored results on Noometry and Step 3.5 Flash has 19.