Step 5 Preview

By StepFun · released 2026-09-18 (12 days) · snapshot 2026-09-30

Off the frontierOn frontier at launch dayproprietary
Intelligence index
43.70
Cost to run the AA index
$929
Price per 1M tokens (in / out)
$1.00 / $2.70
End-to-end time
32.3s
Better than…

Is Step 5 Preview on the Pareto frontier, and what beats it?

Off the frontier

No. GPT-6.1 Sol (xhigh) is at least as smart for the same money or less.

Next step up: GPT-6.1 Sol (max), smarter but dearer.

Step 5 Preview vs GPT-6.1 Sol (xhigh)Step 5 Preview vs GPT-6.1 Sol (max)
#ModelCreatorIntelligenceTotal costE2E totalOpenReleasedFrontier
1Step 5 PreviewStepFun43.70$92932.3s2026-09-18
2GPT-6.1 Sol (xhigh)OpenAI51.00$66277.0s2026-09-29Frontier
3GPT-6.1 Sol (max)OpenAI51.80$1082255.2s2026-09-29Frontier

Open in the picker Daily check URL: 200 on the frontier, 409 once beaten

At a glance

Step 5 Preview today

Time to first answer26.4s
Output speed85 tok/s
Input modalitiesText, Image
0204060$10$20$50$100$200$500$1k$2k$5k$10k$20kCost to run the Artificial Analysis Intelligence Index (USD, log scale)Intelligence IndexApodex 1.1 (Apodex): intelligence 26.40, $939Celeris-1 (Celeris): intelligence 6.30, $60.0Claude 4.5 Haiku (Reasoning) (Anthropic): intelligence 16.90, $594Claude 4.5 Sonnet (Reasoning) (Anthropic): intelligence 20.70, $1501Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) (Anthropic): intelligence 49.60, $11161Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) (Anthropic): intelligence 51.20, $5242Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback) (Anthropic): intelligence 46.80, $3158Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) (Anthropic): intelligence 53.40, $13129Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) (Anthropic): intelligence 48.90, $3983Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) (Anthropic): intelligence 53.20, $9063Claude Opus 4.8 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 41.80, $6874Claude Opus 5 (Adaptive Reasoning, High Effort) (Anthropic): intelligence 48.10, $4332Claude Opus 5 (Adaptive Reasoning, Low Effort) (Anthropic): intelligence 39.40, $1561Claude Opus 5 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 50.80, $7275Claude Opus 5 (Adaptive Reasoning, Medium Effort) (Anthropic): intelligence 44.80, $2732Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) (Anthropic): intelligence 49.70, $5868Claude Opus 5.5 (Adaptive Reasoning, Low Effort, Default Fallback) (Anthropic): intelligence 42.30, $860Claude Opus 5.5 (Adaptive Reasoning, Medium Effort, Default Fallback) (Anthropic): intelligence 51.20, $1627Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 30.10, $4925Claude Sonnet 5 (Adaptive Reasoning, High Effort) (Anthropic): intelligence 31.70, $2086Claude Sonnet 5 (Adaptive Reasoning, Low Effort) (Anthropic): intelligence 24.30, $653Claude Sonnet 5 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 38.20, $6998Claude Sonnet 5 (Adaptive Reasoning, Medium Effort) (Anthropic): intelligence 28.10, $1218Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort) (Anthropic): intelligence 34.40, $3255Claude Sonnet 5.5 (Adaptive Reasoning, High Effort, Default Fallback) (Anthropic): intelligence 46.70, $1176Claude Sonnet 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) (Anthropic): intelligence 56.00, $8977Claude Sonnet 5.5 (Adaptive Reasoning, Medium Effort, Default Fallback) (Anthropic): intelligence 40.70, $701Claude Sonnet 5.5 (Adaptive Reasoning, Xhigh Effort, Default Fallback) (Anthropic): intelligence 51.90, $2738DeepSeek R1 (Jan '25) (DeepSeek): intelligence 11.40, $461DeepSeek V3 (Dec '24) (DeepSeek): intelligence 8.50, $25.4DeepSeek V3 0324 (DeepSeek): intelligence 9.70, $19.2DeepSeek V4 Flash 0420 (Reasoning, High Effort) (DeepSeek): intelligence 24.40, $144DeepSeek V4 Flash 0420 (Reasoning, Max Effort) (DeepSeek): intelligence 24.20, $179DeepSeek V4 Flash 0731 (Reasoning, Max Effort) (DeepSeek): intelligence 34.30, $474DeepSeek V4 Flash Vision (Reasoning, Max Effort) (DeepSeek): intelligence 34.80, $445DeepSeek V4 Pro 0424 (Reasoning, Max Effort) (DeepSeek): intelligence 30.40, $277DeepSeek V4 Pro 0813 (Non-reasoning) (DeepSeek): intelligence 20.40, $566DeepSeek V4 Pro 0813 (Reasoning, Max Effort) (DeepSeek): intelligence 36.00, $1122DeepSeek V4.1 Flash (Non-Reasoning) (DeepSeek): intelligence 24.70, $156DeepSeek V4.1 Flash (Reasoning, Max Effort) (DeepSeek): intelligence 39.50, $477Gemini 2.5 Pro (Google): intelligence 16.10, $707Gemini 3.1 Flash-Lite (Google): intelligence 15.60, $119Gemini 3.1 Pro Preview (Google): intelligence 29.70, $1310Gemini 3.5 Flash (high) (Google): intelligence 32.60, $2172Gemini 3.5 Flash-Lite (Google): intelligence 22.20, $267Gemini 3.6 Flash (high) (Google): intelligence 34.00, $1037Gemini 3.7 Flash (high) (Google): intelligence 39.10, $1084Gemini 3.8 Flash (high) (Google): intelligence 40.90, $1623Gemini 3.8 Flash (medium) (Google): intelligence 39.80, $1100GLM 5.3 Flash (Z AI): intelligence 41.80, $280GLM-5.1 (Reasoning) (Z AI): intelligence 26.10, $1487GLM-5.2 (max) (Z AI): intelligence 33.70, $2097GLM-5.3 (low) (Z AI): intelligence 34.30, $1013GLM-5.3 (max) (Z AI): intelligence 44.80, $2503GPT-5 mini (high) (OpenAI): intelligence 16.80, $189GPT-5.4 mini (xhigh) (OpenAI): intelligence 24.10, $1524GPT-5.4 nano (xhigh) (OpenAI): intelligence 20.70, $331GPT-5.5 (high) (OpenAI): intelligence 37.00, $3411GPT-5.5 (medium) (OpenAI): intelligence 33.80, $2095GPT-5.5 (xhigh) (OpenAI): intelligence 38.40, $5294GPT-5.5 Instant (June 2026) (OpenAI): intelligence 26.00, $1184GPT-5.6 Luna (high) (OpenAI): intelligence 32.10, $108GPT-5.6 Luna (max) (OpenAI): intelligence 37.30, $320GPT-5.6 Luna (medium) (OpenAI): intelligence 25.00, $44.3GPT-5.6 Luna (Non-reasoning) (OpenAI): intelligence 15.50, $23.0GPT-5.6 Luna (xhigh) (OpenAI): intelligence 34.60, $179GPT-5.6 Sol (high) (OpenAI): intelligence 42.30, $1487GPT-5.6 Sol (low) (OpenAI): intelligence 33.50, $637GPT-5.6 Sol (max) (OpenAI): intelligence 47.00, $3465GPT-5.6 Sol (medium) (OpenAI): intelligence 39.20, $997GPT-5.6 Sol (xhigh) (OpenAI): intelligence 44.00, $2082GPT-5.6 Terra (high) (OpenAI): intelligence 34.20, $768GPT-5.6 Terra (low) (OpenAI): intelligence 27.50, $350GPT-5.6 Terra (max) (OpenAI): intelligence 42.10, $2501GPT-5.6 Terra (medium) (OpenAI): intelligence 30.10, $447GPT-5.6 Terra (Non-reasoning) (OpenAI): intelligence 20.80, $256GPT-5.6 Terra (xhigh) (OpenAI): intelligence 38.00, $1187GPT-6 Astra (high) (OpenAI): intelligence 50.90, $2925GPT-6 Astra (low) (OpenAI): intelligence 45.80, $1537GPT-6 Astra (max) (OpenAI): intelligence 52.70, $5324GPT-6 Astra (medium) (OpenAI): intelligence 49.60, $2434GPT-6 Astra (xhigh) (OpenAI): intelligence 52.40, $3803GPT-6 Luna (max) (OpenAI): intelligence 37.30, $122GPT-6 Luna (Non-reasoning) (OpenAI): intelligence 18.30, $19.4GPT-6 Sol (high) (OpenAI): intelligence 42.80, $610GPT-6 Sol (low) (OpenAI): intelligence 33.90, $268GPT-6 Sol (max) (OpenAI): intelligence 47.50, $1546GPT-6 Sol (medium) (OpenAI): intelligence 39.80, $416GPT-6 Sol (Non-reasoning) (OpenAI): intelligence 28.10, $449GPT-6 Sol (xhigh) (OpenAI): intelligence 44.10, $860GPT-6.1 Sol (low) (OpenAI): intelligence 42.10, $250gpt-oss-120b (high) (OpenAI): intelligence 11.60, $122gpt-oss-20b (high) (OpenAI): intelligence 9.00, $31.4Granite 4.2 3B (IBM): intelligence 9.10, $14.9Granite 4.2 8B (IBM): intelligence 11.10, $53.1Grok 4.3 (high) (SpaceXAI): intelligence 24.90, $363Grok 4.3 (Non-reasoning) (SpaceXAI): intelligence 14.00, $250Grok 4.5 (high) (SpaceXAI): intelligence 38.80, $1404Grok 4.6 (high) (SpaceXAI): intelligence 44.30, $2352Grok 4.6 (low) (SpaceXAI): intelligence 35.10, $761Grok 4.6 (medium) (SpaceXAI): intelligence 42.80, $1937Grok 4.6 (xhigh) (SpaceXAI): intelligence 44.20, $2830Grok 4.7 (high) (SpaceXAI): intelligence 46.30, $3881Grok 4.7 (xhigh) (SpaceXAI): intelligence 46.40, $4967Hy3 (Tencent): intelligence 25.30, $150Inkling Small (Thinking Machines): intelligence 25.70, $241Kimi K2.6 (Kimi): intelligence 27.00, $1528Kimi K2.7 Code (Kimi): intelligence 25.80, $993Kimi K3 (low) (Kimi): intelligence 30.10, $1190Kimi K3 (max) (Kimi): intelligence 43.60, $3658LongCat 2.0 (LongCat): intelligence 19.10, $221Mercury 2.5 (Inception): intelligence 12.30, $218MiMo-V2.5-Pro (Xiaomi): intelligence 26.00, $138MiniMax-M3 (MiniMax): intelligence 29.20, $538Ministral 3 14B (Mistral): intelligence 6.00, $37.6Ministral 3 3B (Mistral): intelligence 4.80, $12.2Ministral 3 8B (Mistral): intelligence 5.50, $27.5Mistral Large 3 (Mistral): intelligence 9.30, $46.4Mistral Medium 3.5 (Mistral): intelligence 14.20, $1264Mistral Small 3.1 (Mistral): intelligence 7.10, $38.1Mistral Small 4 (Reasoning) (Mistral): intelligence 11.30, $42.5Muse Glimmer (high) (Meta): intelligence 17.50, $149Muse Spark 1.1 (xhigh) (Meta): intelligence 33.70, $1737Muse Spark 1.2 (xhigh) (Meta): intelligence 39.60, $1385Muse Spark 1.3 (max) (Meta): intelligence 48.10, $2000Muse Spark 1.3 (xhigh) (Meta): intelligence 45.10, $1655Nemotron 3 Super 120B A12B (Reasoning) (NVIDIA): intelligence 12.80, $1319Nemotron 3 Ultra 550B A55B (Reasoning) (NVIDIA): intelligence 22.90, $805Nemotron 3.5 Lightning (NVIDIA): intelligence 12.90, $114NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) (NVIDIA): intelligence 8.90, $63.7Quasar 438B (max, based on GLM-5.2) (Multiverse Computing): intelligence 26.70, $2504Qwen3 235B A22B 2507 (Reasoning) (Alibaba): intelligence 12.70, $235Qwen3 30B A3B 2507 (Reasoning) (Alibaba): intelligence 9.80, $188Qwen3 Coder Next (Alibaba): intelligence 9.20, $476Qwen3.5 122B A10B (Reasoning) (Alibaba): intelligence 15.60, $661Qwen3.5 397B A17B (Reasoning) (Alibaba): intelligence 18.40, $941Qwen3.5 9B (Reasoning) (Alibaba): intelligence 11.20, $329Qwen3.6 27B (Reasoning) (Alibaba): intelligence 21.40, $1145Qwen3.6 35B A3B (Reasoning) (Alibaba): intelligence 18.20, $896Qwen3.7 Max (Alibaba): intelligence 29.50, $2277Qwen3.7 Plus (Alibaba): intelligence 25.20, $428Qwen3.8 2.4T A95B (Alibaba): intelligence 39.90, $2633Qwen3.8 27B (low) (Alibaba): intelligence 26.20, $1060Qwen3.8 27B (medium) (Alibaba): intelligence 27.60, $1165Qwen3.8 27B (Non-reasoning) (Alibaba): intelligence 20.20, $2022Qwen3.8 27B (xhigh) (Alibaba): intelligence 33.70, $1336Qwen3.8 Max (Alibaba): intelligence 40.20, $3172Qwen3.8 Max (0902) (Alibaba): intelligence 45.40, $4935Qwen3.8-Flash-Next (Alibaba): intelligence 39.80, $363Ring-2.6-1T (InclusionAI): intelligence 16.60, $576Solar Pro 3 (Upstage): intelligence 7.80, $159Trinity Large Thinking (Arcee AI): intelligence 10.80, $233GPT-6 Luna (low) (OpenAI): intelligence 20.90, $10.6GPT-5.6 Luna (low) (OpenAI): intelligence 21.00, $27.0GPT-6 Luna (medium) (OpenAI): intelligence 29.50, $31.4GPT-6 Luna (high) (OpenAI): intelligence 32.10, $47.9GPT-6 Luna (xhigh) (OpenAI): intelligence 33.90, $66.6MiMo-V2.6-Flash (Xiaomi): intelligence 37.90, $109MiMo-V2.6-Pro (Xiaomi): intelligence 46.30, $207GPT-6.1 Sol (medium) (OpenAI): intelligence 47.80, $361GPT-6.1 Sol (high) (OpenAI): intelligence 50.20, $521GPT-6.1 Sol (xhigh) (OpenAI): intelligence 51.00, $662GPT-6.1 Sol (max) (OpenAI): intelligence 51.80, $1082Claude Opus 5.5 (Adaptive Reasoning, High Effort, Default Fallback) (Anthropic): intelligence 53.60, $2172Claude Opus 5.5 (Adaptive Reasoning, Xhigh Effort, Default Fallback) (Anthropic): intelligence 56.00, $4057Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) (Anthropic): intelligence 57.60, $8708Step 5 Preview (StepFun): intelligence 43.70, $929Step 5 PreviewGPT-6 Luna (low)GPT-6.1 Sol (xhigh)GPT-6.1 Sol (max)Claude Opus 5.5 (max)
Step 5 PreviewSame release, other effort levelsToday's Pareto frontierEvery other model
Up and left is better. Dotted line: today's Pareto frontier. Hover a point for its numbers, click it for its page.

Intelligence: Artificial Analysis Intelligence Index. Cost: total cost to run that benchmark, so reasoning tokens count. API id: step-5.

Matchup

Compare head to head

Suggested: what beats it, its effort levels, and frontier models near its price.

Compare Step 5 Preview with measured models

Release

What the Step 5 Preview release did to the frontier on 2026-09-18

1 of 1 effort levels reached the frontier at launch (Step 5 Preview). Impact 0.41, #28 of 37 releases this past year.

Release page: launch-day frontier and every effort level

Deciding

Should you use Step 5 Preview?

Should you use Step 5 Preview?

Only when it fits your setup: GPT-6.1 Sol (xhigh) is at least as smart for the same money or less, so Step 5 Preview is off the value frontier today.

What is better and cheaper than Step 5 Preview?

GPT-6.1 Sol (xhigh) — at least as smart for the same money or less.

How much does Step 5 Preview cost?

$1.00/$2.70 in/out per 1M tokens, and $929 to run the full Artificial Analysis Intelligence Index, the honest all-in number for reasoning models.

Is Step 5 Preview on the Pareto frontier?

No: GPT-6.1 Sol (xhigh) beats it on intelligence, cost, or both.

Frontier membership by axis pair

Other axis pairs where Step 5 Preview is Pareto-optimal.

Y axisX axisOn the frontier
AA intelligence indexAA total cost to run (USD)
AA intelligence indexEnd-to-end total seconds
AA intelligence indexOutput speed tokens/sec
Agents: Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/models/step-5 · Markdown: /api/v1/models/step-5?format=md · llms.txt · agent skill