Claude Haiku 4.5 vs Step 5 Preview
Anthropic · US | StepFun · China · Updated June 2026
Quick verdict
Pick Claude Haiku 4.5 for fastest claude model or near-frontier coding for its tier — 73.3% on swe-bench verified. Pick Step 5 Preview for sparse moe architecture — 600b total parameters, only 27b active per token — built for long-horizon agentic work across large codebases and documents or 1m-token context window (64k max output) at $1/$2.70 per million tokens, among the cheapest frontier-scale context windows available.
Claude Haiku 4.5 (Anthropic, US) and Step 5 Preview (StepFun, China) line up two different AI ecosystems against each other — a comparison that is as much about cost philosophy and openness as raw capability. Claude Haiku 4.5 is anthropic's fastest, most compact model — built for speed and volume. Step 5 Preview is stepFun's September 20, 2026 flagship preview — a 600B-parameter sparse MoE model (27B active) with a 1M-token context window at $1/$2.70 per million tokens, open weights promised for mid-October. Their biggest split is context window, and the breakdown below shows exactly how that plays out for your workload.
Key differences at a glance
- ▸Context window: Step 5 Preview holds 5× more — 1M tokens (~1,500 pages) vs 200K (~300 pages). But effective recall usually fades long before the advertised ceiling, so the bigger number only helps if the model reasons over it.
- ▸Recency: Step 5 Preview is the newer model by about 11 months (released September 20, 2026), usually meaning fresher training data and capabilities.
- ▸Ecosystem: this is a US-vs-China matchup — they differ in pricing philosophy, data-residency options, and tooling ecosystems, not only benchmarks.
Side-by-side specs
| Spec | Claude Haiku 4.5 | Step 5 Preview |
|---|---|---|
| Provider | Anthropic (US) | StepFun (China) |
| Released | October 15, 2025 | September 20, 2026 |
| Context window | 200K (~300 pages) | 1M tokens (~1,500 pages) |
| Price (in/out) | $1/$5 per 1M tokens | $1/$2.7 per 1M tokens |
| Open weight? | No — API only | No — API only |
| Modalities | text, image, code | text, image, code |
| SWE-Bench Verified | 73.3% | Not published |
| MRCR v2 @ 1M | Not published | Not published |
Who wins what
Fastest Claude model
Claude Haiku 4.5
Step 5 Preview is comparatively weak here — trails closed frontier models on coding — 67.7% on DeepSWE v1.1 vs GPT-6 Astra's 74.1% and Claude Opus 5's 74.0%
Near-frontier coding for its tier — 73.3% on SWE-Bench Verified
Claude Haiku 4.5
Claude Haiku 4.5 lists near-frontier coding for its tier — 73.3% on SWE-Bench Verified among its strengths; Step 5 Preview does not.
Low-latency, high-volume API calls
Claude Haiku 4.5
Claude Haiku 4.5 lists low-latency, high-volume API calls among its strengths; Step 5 Preview does not.
Sparse MoE architecture — 600B total parameters, only 27B active per token — built for long-horizon agentic work across large codebases and documents
Step 5 Preview
Its 1M tokens window holds about 5× more than Claude Haiku 4.5's 200K in a single prompt.
1M-token context window (64K max output) at $1/$2.70 per million tokens, among the cheapest frontier-scale context windows available
Step 5 Preview
Claude Haiku 4.5 is comparatively weak here — smallest context in the family (200K)
67.7% on DeepSWE v1.1, ahead of open-weight rivals Kimi K3 (67.5%) and GLM-5.3 (66.9%)
Step 5 Preview
StepFun's September 20, 2026 flagship preview — a 600B-parameter sparse MoE model (27B active) with a 1M-token context window at $1/$2.70 per million tokens, open weights promised for mid-October — and it carries the larger 1M tokens context.
Largest single-prompt input
Step 5 Preview
Its 1M tokens window is about 5× larger than Claude Haiku 4.5's 200K, fitting roughly 1,500 pages in one prompt.
Which should you pick?
Someone analysing very long documents or codebases
→ Step 5 Preview
Larger 1M tokens window fits more in one prompt.
Anyone whose priority is fastest claude model
→ Claude Haiku 4.5
It is specifically built for that.
Anyone whose priority is sparse moe architecture — 600b total parameters, only 27b active per token — built for long-horizon agentic work across large codebases and documents
→ Step 5 Preview
That is its strongest area.
An enterprise with regional data-residency rules
→ Claude Haiku 4.5 or Step 5 Preview
Origin (US vs China) affects where data is processed and which compliance regime applies — check the provider's terms for your region.
Claude Haiku 4.5: where it fits
Anthropic's fastest, most compact model — built for speed and volume. Released October 15, 2025 by Anthropic, it is built for fastest Claude model, near-frontier coding for its tier — 73.3% on SWE-Bench Verified, low-latency, high-volume API calls, and cheapest Claude tier at $1/$5.
Its trade-offs are real: smallest context in the family (200K), and not for deep reasoning. At $1 in / $5 out per million tokens, it sits in the budget price band.
Step 5 Preview: where it fits
StepFun's September 20, 2026 flagship preview — a 600B-parameter sparse MoE model (27B active) with a 1M-token context window at $1/$2.70 per million tokens, open weights promised for mid-October. Released September 20, 2026 by StepFun, it is built for sparse MoE architecture — 600B total parameters, only 27B active per token — built for long-horizon agentic work across large codebases and documents, 1M-token context window (64K max output) at $1/$2.70 per million tokens, among the cheapest frontier-scale context windows available, 67.7% on DeepSWE v1.1, ahead of open-weight rivals Kimi K3 (67.5%) and GLM-5.3 (66.9%), and scores 44 on the Artificial Analysis Intelligence Index, ranking 27th of 653 models tracked.
Its trade-offs: trails closed frontier models on coding — 67.7% on DeepSWE v1.1 vs GPT-6 Astra's 74.1% and Claude Opus 5's 74.0%, open weights were promised for October 15, 2026 but hadn't shipped as of this writing — API-only for now despite the open-weight roadmap, a 'Preview' release, not StepFun's finished flagship — positioning and pricing may change once weights ship, and stepCodeBench and DeepSWE figures are StepFun's own reported numbers, not yet independently reproduced at scale. At $1 in / $2.7 out per million tokens, it sits in the budget price band.
The bottom line for this matchup
This is less "which is smarter" and more "which ecosystem fits." Claude Haiku 4.5 (US) and Step 5 Preview (China) differ on pricing philosophy, data-residency, and tooling as much as on raw scores. The pragmatic move is to run one real task through both and judge the outputs against your own constraints — including where your data is allowed to be processed.
Want both Claude Haiku 4.5 and Step 5 Preview without two subscriptions? LumiChats gives you these plus 40+ models under one ₹69/day pass (about $1/day) — draft with one, cross-check with the other.
See pricingFrequently asked questions
Is Claude Haiku 4.5 or Step 5 Preview better for coding?
Public SWE-Bench figures are not available for Step 5 Preview, so the honest test is your own repository — run an identical real bug through both. By design, Claude Haiku 4.5 leans toward fastest claude model while Step 5 Preview leans toward sparse moe architecture — 600b total parameters, only 27b active per token — built for long-horizon agentic work across large codebases and documents, and that positioning usually predicts which feels better on your codebase.
Which is cheaper, Claude Haiku 4.5 or Step 5 Preview?
They are priced almost identically, so cost will not decide between them.
Which has the bigger context window?
Step 5 Preview — 1M tokens vs 200K, about 5× larger. Useful only if the model actually reasons over the full window, which not all do.
Can I use both Claude Haiku 4.5 and Step 5 Preview together?
Yes — a multi-model platform like LumiChats gives you Claude Haiku 4.5, Step 5 Preview and 40+ others under one ₹69/day pass (about $1/day), so you can draft with one and cross-check with the other instead of buying two subscriptions.
Which is newer, Claude Haiku 4.5 or Step 5 Preview?
Step 5 Preview — released September 20, 2026, about 11 months after Claude Haiku 4.5.
Related comparisons
Specifications and benchmarks reflect publicly reported figures as of June 2026 and may change as providers release updates. Always verify on your own workload.