StepFun's Step 5 Preview matches Kimi K3 at ~2.8x lower cost per task
ArtificialAnlys · x · 2026-09-22
Artificial Analysis estimates that StepFun's new flagship Step 5 Preview (600B total / 27B active parameters, succeeding Step 3.7 Flash) scores 44 on the Intelligence Index — level with Kimi K3 (max) — while costing only $0.72 per task, roughly 2.8x cheaper than Kimi K3's $2.00.
The gap is pricing, not token usage: both models use 164–170M output tokens to run the Index, but Step 5 Preview is priced at $1/$2.70 per 1M input/output tokens vs $3/$15 for Kimi K3. GLM-5.3 (max) costs $2.01 per task for one extra point. MiMo-V2.6-Pro is the only model scoring higher (46) at lower cost ($0.13). Note: Step 5 Preview trails peers on agentic evaluations.
More from Models
- xAI's Grok hype cycle repeats: 4.7 delayed, no frontier model beats ChatGPT or Claude — flowersslop · 2026-09-22
- xAI fixes SDK bug dropping reasoning content, significantly boosting Grok 4.7 — ns123abc · 2026-09-22
- TTS leaderboard: xAI hits 87.6% pronunciation accuracy, Kokoro 82M fastest at 242 chars/s — ArtificialAnlys · 2026-09-22
- xAI's Post-Launch SDK Update Pushes It to #10 on the Vals Index — teortaxesTex · 2026-09-22
- Grok 4.7 looks pricier than before, now costlier than Astra on Artificial Analysis — steipete · 2026-09-22
- Which LLM is most encyclopedic on 8GB VRAM + 64GB RAM? — Mangleus · 2026-09-22