Step 5 Preview
StepFunProprietaryReleased Sep 18, 2026
On the arena.ai Agents leaderboard, Step 5 Preview has a Net Improvement of -3.74% (—, rank #40, 95% interval -5.15%–-2.34%). Step 5 Preview ranks #84 on the LMArena Text leaderboard (style control) with an Arena score of 1448.5 (±6.3, 12,345 votes). Artificial Analysis Intelligence Index: 28.5. API price: $1.00 input / $2.70 output per 1M tokens. Context: 1M. Nearest rivals: Kimi K2.5 (1449.5) above and MiMo V2.6 Flash (1448.4) below.
Quick answer
- Agents leaderboard
- #40 of 56 · -3.74% success
- Text (LMArena)
- #84 · 1448.5
- 17NAS overall
- #39 of 55
- API price (in / out)
- $1.00 / $2.70 per 1M tokens
- Closest rival
- Kimi K2.5 · 1449.5
- Cheaper alternative
- Qwen3.7 Plus: Text 1455.3, output $1.60 (41% cheaper)
- Data updated
- Nov 15, 2026 · synced daily
Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter
Key metrics
Models ranked nearby
Sorted by Arena score| # | Model | Arena | Intel. | Output $ | tok/s |
|---|---|---|---|---|---|
| 54 |  Hy3 | 1454.0 | – | $0.56 | 71 |
| 55 |  GPT-5.6 Luna | 1453.2 | 23.8 | $1.20 | 127 |
| 56 |  Kimi K2.5 | 1449.5 | – | $3.00 | 54 |
| 57 | SStep 5 PreviewThis page | 1448.5 | 28.5 | $2.70 | 118 |
| 58 |  MiMo V2.6 Flash | 1448.4 | 26.7 | $0.28 | 34 |
| 59 |  Gemini 2.5 Pro | 1445.5 | – | $10.00 | 116 |
| 60 |  Grok 4.7 | 1443.3 | 35.5 | $6.00 | 74 |
Pricing
- Official API list price: $1.00 input and $2.70 output per 1M tokens (USD).
- At a typical 3:1 input-to-output ratio, the blended cost is about $1.43 per 1M tokens.
- Among 95 models with an output price, it is the #36 cheapest.
- The cheapest OpenRouter route charges $1.00 input / $2.70 output.
Sources & methodology
On the arena.ai Agents leaderboard, Step 5 Preview has a Net Improvement of -3.74% (—, rank #40, 95% interval -5.15%–-2.34%).
- arena.ai Agents — agent Net Improvement, 95% interval, sessions and rank (via the public Hugging Face dataset, CC BY 4.0).
- LMArena — Arena score (Text leaderboard with style control, the same view arena.ai shows by default), 95% confidence interval, votes and rank.
- Artificial Analysis — Intelligence / Coding Index, output speed and latency (medians).
- OpenRouter — Route prices, context length and listing date.
Arena scores use the Bradley-Terry model (not Elo); ± is the half-width of the 95% confidence interval. Latency figures are Artificial Analysis medians; for reasoning models, time to first token and to first answer include the thinking phase. Data syncs daily at 14:20 (UTC+8).
Limits: Arena scores come from anonymous users voting on chats, so they reflect conversational preference rather than agent or coding benchmarks; with few votes the interval is wide and ranks move. Prices and speeds change as providers update them.
FAQ
How does Step 5 Preview do on the Agents leaderboard?
On the arena.ai Agents leaderboard, Step 5 Preview has a Net Improvement of -3.74% (—, rank #40, 95% interval -5.15%–-2.34%).
What rank is Step 5 Preview right now?
Step 5 Preview is #57 on our leaderboard of 80 scored models (Arena score 1448.5); its raw LMArena rank is #84.
What does Step 5 Preview cost via API?
Input $1.00 / output $2.70 per 1M tokens (official API list price; the OpenRouter quote when no official price is available). At a typical 3:1 input-to-output ratio, the blended cost is about $1.43 per 1M tokens.
How fast is Step 5 Preview?
Artificial Analysis measures about 118 output tokens per second, 0.8 s to the first token and 18 s to the first answer token; it is a reasoning model, so latency includes thinking time.
How close is Step 5 Preview to the models ranked next to it?
It is 1.0 points behind Kimi K2.5 (1449.5) and 0.1 points ahead of MiMo V2.6 Flash (1448.4); gaps smaller than the confidence interval (±6.3) are not significant.
Is Step 5 Preview open source? Can it be self-hosted?
Step 5 Preview is a closed model, available only via official APIs or hosted services.
What is the context length of Step 5 Preview?
1M (1,000,000 tokens, as listed on OpenRouter).
Step 5 Preview vs Kimi K2.5: which is better?
On the LMArena Text leaderboard Step 5 Preview scores 1448.5 and Kimi K2.5 scores 1449.5 (a 1.0-point gap, within the confidence interval, so not significant). On price, Step 5 Preview costs $2.70 and Kimi K2.5 $3.00 per 1M output tokens.
Is there a cheaper model close to Step 5 Preview?
Yes: Qwen3.7 Plus scores 1455.3 on the Text leaderboard (Step 5 Preview: 1448.5) and costs $1.60 per 1M output tokens, about 41% less than Step 5 Preview ($2.70).
Where does the Step 5 Preview data come from and how often is it updated?
Agent success rates come from the arena.ai Agents leaderboard (CC BY 4.0), Text scores from LMArena, intelligence and speed from Artificial Analysis, and prices and context from official list prices and OpenRouter. 17NAS syncs daily at 06:20 UTC; this page was last updated Nov 15, 2026.