StepFun flash model for efficient multimodal reasoning, coding, and tool use
Step 3.5 Flash 2603 by stepfun is offered by 6 providers on this page. Public prices are shown for 4 of them. Its official list price is $0.10 per 1M input tokens and $0.30 per 1M output tokens. It also has 2 covered by a paid subscription; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.
Artificial Analysis rates it 17 on the Intelligence Index. Against its lowest blended price of $0.15 per 1M, that is roughly 113 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 114 tokens per second, with 3.42s to the first token.
The context window is 256,000 tokens at the reference host, but hosts report different limits, from 256,000 to 262,144, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2025-01.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| EmpirioLabs AI step-3-5-flash-2603 | Gateway | $0.10 | $0.30 | $0.02 | — | 256,000 | 131,072 | |
| NanoGPT | Gateway | $0.10 | $0.30 | $0.05 | — | 262,144 ⚠ | 65,536 | |
| StepFun (Global)Official | First-party | $0.10 | $0.30 | $0.02 | — | 256,000 | 256,000 | |
| StepFun (China)Official | First-party | $0.10 | $0.30 | $0.02 | — | 256,000 | 256,000 | |
| StepFun Step Plan (Global) | Gateway | — | — | — | — | 256,000 | 256,000 | hostTable.undisclosed |
| StepFun Step Plan (China) | Gateway | — | — | — | — | 256,000 | 256,000 | hostTable.undisclosed |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Interleaved thinking (reasoning between tool calls) is declared by 4 of 6 providers.
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| Artificial Analysis Coding Index | — | 34.6 | index | Source ↗ |
| SciCode | — | 38.5 | percent correct | Source ↗ |
| Terminal-Bench Hard | — | 32.6 | success rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and harness, so rows are not comparable with one another. The Intelligence Index above draws on a different, newer set of evaluations.