Step 3.5 Flash 0202
Step 3.5 Flash 0202 is a StepFun open-weights reasoning model, released 2 Feb 2026. Its best configuration ranks #360 of 784 on the BenchLeader Index at 50.8 ±5.9, in the upper half. The ± is the point: 349 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (64) and lowest in knowledge (43). At $0.150 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 145 tokens per second puts it in the fastest quarter, with a first answer in 17.1 s. Last measured 2 Feb 2026.stale: no new result in six months
- Blended price
- $0.150/M
- $0.100 in · $0.300 out
- Output speed
- 145 tok/s
- measured by Artificial Analysis
- First answer
- 17 s
- Context
- 256k
- Full answer
- 21 s
- median, reasoning included
- Cost per run
- –
- Released
- 2 Feb 2026
- Overall index51
- Reasoning45
- Agents & tools58
- Knowledge43
- Instruction following64
- Long context50
- Composite48
Versions
StepFun has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Step 3.7 Flash | 29 May 2026 | 51.5 | #342 |
| Step 3.5 Flash | 2 Apr 2026 | 53.7 | #275 |
| Step 3.5 Flash 0202this page | 2 Feb 2026 | 50.8 | #360 |
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source |
|---|---|---|
| GPQA Diamond (AA)not in index | 83.1%#161 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 21.1%#209 | Artificial Analysis |
| CritPt | 2.5%#196 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source |
|---|---|---|
| Terminal-Bench Hard | 27.3%#125 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 94.4%#30 | Artificial Analysis |
Knowledge
| Benchmark | not stated | Source |
|---|---|---|
| AA-Omniscience | -41.8#329 | Artificial Analysis |
| AA-Omniscience: accuracynot in index | 23.6%#254 | Artificial Analysis |
| AA-Omniscience: non-hallucinationnot in index | 14.3%#348 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source |
|---|---|---|
| IFBench | 64.6%#107 | Artificial Analysis |
Long context
| Benchmark | not stated | Source |
|---|---|---|
| AA-LCR | 50.0%#320 | Artificial Analysis |
Composite
| Benchmark | not stated | Source |
|---|---|---|
| AA Intelligence Index v4.3.2 | 16.6#269 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | – | 19.2 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0014 | – | 21.2 s |
| Code edit | 6,000 / 1,500 | $0.0011 | – | 27.4 s |
| Agentic coding session | 60,000 / 4,000 | $0.0072 | – | 44.7 s |
| Structured extraction | 2,000 / 200 | $0.0003 | – | 18.5 s |
See also
Data as of 22 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Step 3.5 Flash 0202: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/step-3-5-flash-0202, data as of 22 Oct 2026.