- Home
- Image model comparison
- Hy Image 3.0 vs. Krea 2 Raw
TencentKrea.aiImage model matchup
Hy Image 3.0vs.Krea 2 Raw
Which one is better?
We have not tested Hy Image 3.0 or Krea 2 Raw ourselves yet. Neither is in the arena. Neither has an API price.
Test images from GraduallyBench Image v2
Same task, the first image of each.
No image from the four fixed tasks for Hy Image 3.0 yet.
No image from the four fixed tasks for Krea 2 Raw yet.
Result for “Typography and layout”:no comparison
What the rating looks at: Legible text, hierarchy, and composition.
“Ahead”: the 95 % ranges of both values do not overlap. “Slightly ahead”: a higher value, but the ranges overlap.
Missing image run for Hy Image 3.0
No enabled Tencent API key was available; no card authorization or balance hold was initiated. The provider announces a November 12, 2026 shutdown for the current hy-image-v3 model ID. An older route with a different alias has already shut down. Last checked November 2, 2026. Before testing, check the current route, its status, and authorized access.
DocumentationMissing image run for Krea 2 Raw
The official demo offers the Raw checkpoint. On November 2, 2026, its default configuration required more allowance than anonymous daily access provided, and a security challenge blocked free sign-up. Local license eligibility remains unresolved. No exact Raw API route through fal.ai, Together or BFL was established on November 3. Krea 2 Large and its “raw” prompt option are not substitutes for the Raw model. Resolve regular account access and license eligibility, then test the exact Raw checkpoint with sufficient allowance.
DocumentationSubscription and API
Hy Image 3.0
- Subscription
- n/a No plan in Germany names exactly this model.
- Via API
- No API price known (access: API).
- Weights
- Unknown.
Krea 2 Raw
- Subscription
- n/a No plan in Germany names exactly this model.
- Via API
- No API price known (access: local).
- Weights
- Open weights.
Subscriptions only count plans that name exactly this model. Subscription prices include VAT, API prices are in dollars.
GraduallyBench Image v2 from Nov 7, 2026 · Subscription database from Sep 29, 2026 (45 days old)
All data side by side
The better value is bold. For ratings only when the gap is clear.
| Kennzahl | Hy Image 3.0 | Krea 2 Raw |
|---|---|---|
| Ratings | ||
| Arena ratingfrom votes by users, higher is better | n/a | n/a |
| Arena rank | n/a | n/a |
| Arena votesmore votes, a safer rating | n/a | n/a |
| GraduallyBench Image v2our own benchmark, 0 to 100 | n/a | n/a |
| GraduallyBench Image v2 by task | ||
| Typography and layoutLegible text, hierarchy, and composition | n/a | n/a |
| Product photographyMaterials, reflections, and fine details | n/a | n/a |
| Character and detailAnatomy, clothing, and spatial consistency | n/a | n/a |
| InfographicNumbers, labels, and visual organization | n/a | n/a |
| API price | ||
| Per imagelower is better | n/a | n/a |
| 100 images | n/a | n/a |
| Profile | ||
| Released | n/a | Jun 22, 2026 |
| Weightsopen means: runs locally | unknown | open |
| Lifecycle | Tencent TokenHub still lists the T2I endpoint as pending shutdown on November 12, 2026; only a short-lived legacy cohort beside Hy Image 3.5 Preview | no notice |
| Access | API | local |
| Parameters | not published | 12 billion |
| Largest output | not consistently published | not consistently published |
| Plans in Germanynaming exactly this model | n/a | n/a |
API prices for one image at 1,024 × 1,024 pixels according to Artificial Analysis.
Sources: Hy Image 3.0 model source, status source (verified Nov 1, 2026) · Krea 2 Raw model source (verified Nov 1, 2026)
Arena ratings by category
From blind votes on Artificial Analysis and LMArena; the sources are named under the card. “Ahead” means the lead is clear.
No shared arena score in the stored snapshot
The stored benchmark database contains no measurement for Hy Image 3.0 and Krea 2 Raw from the same arena and category. Older model versions are not used as substitutes.
More comparisons
Compare Hy Image 3.0 and Krea 2 Raw with the models people compare them with most often.