Best AI image generator for golden hour landscapes
25 models compared (November 2026)
Best for golden hour landscapes: Recraft V4.1 Pro (7.8 out of 10), ahead of FLUX.3 Klein and Recraft V4.1 Utility Pro, from our scores for prompt adherence, realism and film look.
We sent the same prompt, word for word, to the 25 best-ranked AI image models and put every result side by side. One image each, no retouching, no picking the best of several. Each card shows the model, its maker and its benchmark score.
The prompt
Extreme wide shot from a dune crest, slightly above the caravan, the line of camels crossing the centre of the frame on a gentle diagonal, the lead handler near the middle. Twelve dromedaries walk roped nose to tail, laden with woven saddlebags and rolled rugs. The handler, a man in his sixties in an indigo cotton robe and wrapped headcloth, walks with a long stick, gaze on the horizon. The sun sits a hand's width above the horizon behind the caravan, golden at 2800K, throwing impossibly long shadows across the dunes toward camera, each leg a thin dark stripe. Wind lifts a veil of sand off the crests into the light, every ripple in sharp relief, and the open sky gives cool blue fill. IMAX 65mm film, an 80mm spherical prime at f/8, deep focus, Kodak Vision3 50D, nearly invisible grain, amber and cobalt palette, a frame from an epic desert saga. Real fabric weave and camel hair, no CGI look. No text, no logos, no watermarks.
- Shape
- 16:9, or the nearest a model offers
- Made
- Models
- 25
How we ran it and ranked it
Every model got the prompt exactly as written above. Where a model can rewrite or expand a prompt before it draws, that was switched off, so each one worked from the same words. We made one image per model at 16:9; a model that cannot make 16:9 made its nearest shape, and its picture is trimmed slightly at the edges to fill the frame. Where a model drew black cinema bars into its picture, the bars are cut off for display. Nothing was retouched or chosen from several tries. A model that failed was tried once more; if it failed again, its card says so and why.
The order is the models' benchmark rank, not our opinion: their score in the Artificial Analysis text-to-image arena (human preference Elo, checked 8 November 2026), or, for a model Artificial Analysis does not score, the Arena text-to-image leaderboard (checked 6 November 2026). The two boards use different scales, so each score is placed by the share of its own board it beats. A model sold under two names is one model, shown once under both.
Only models whose pictures may be used commercially are included.
The answer at the top and the notes under What we noticed come from our own scores, not the benchmark. Each picture was scored 1 to 10 on prompt adherence, realism and film look, and the scores were checked by an independent cross-review.
The 25 results
Ranked by benchmark. Select a picture to see it full size, then use the arrows or your arrow keys to compare.
-
#1
Nano Banana 2
· Google· AA Elo 1142 -
#2
Muse Image
· Meta· AA Elo 1128 -
#3
Nano Banana Pro
· Google· AA Elo 1119 -
#4
Qwen Image 3.0 Pro
· Alibaba· AA Elo 1095 -
#5
Qwen Image 3.0
· Alibaba· AA Elo 1073 -
#6
FLUX.3 Klein
· Black Forest Labs· AA Elo 1043 -
#7
FIBO 2.0
· Bria· AA Elo 1039 -
#8
Seedream 4.5
· ByteDance· AA Elo 1030 -
#9
FLUX.2 [flex]
· Black Forest Labs· AA Elo 1028 -
#10
Recraft V4.1 Utility
· Recraft· AA Elo 1028 -
#11
Luma Uni-1 Max
· Luma AI· AA Elo 1026 -
#12
Grok Imagine Image
· xAI· AA Elo 1024 -
#13
FLUX.2 [max]
· Black Forest Labs· AA Elo 1021 -
#14
Seedream 5.0 Lite
· ByteDance· AA Elo 1019 -
#15
Recraft V4.1 Utility Pro
· Recraft· AA Elo 1018 -
#16
Wan 2.6
· Alibaba· AA Elo 1018 -
#17
FLUX.2 [pro]
· Black Forest Labs· AA Elo 1003 -
#18
FLUX.2 [dev]
· Black Forest Labs· AA Elo 1000 -
#19
FLUX.2 [dev] Turbo
· Black Forest Labs· AA Elo 996 -
#20
Luma Uni-1
· Luma AI· AA Elo 994 -
#21
Recraft V4.1 Pro
· Recraft· Arena 1132 -
#22
FLUX.2 [dev] Flash
· Black Forest Labs· AA Elo 989 -
#23
HunyuanImage 3.0 Instruct
· Tencent· AA Elo 988 -
#24
Wan 2.7 Image
· Alibaba· AA Elo 988 -
#25
Cosmos 3 Super
· NVIDIA· AA Elo 987
What we noticed
Our read of these pictures. It's one image per model, so treat it as a snapshot, not a verdict.
Strongest: Recraft V4.1 Pro. Weakest: Wan 2.6.
Recraft V4.1 Pro is the closest to the brief: a true extreme wide, the sun a hand above the horizon, thin leg-stripe shadows toward camera, spindrift and a cobalt and amber palette. It shows only 8 camels, and the handler leads the line rather than walking mid-line. Wan 2.6 went the other way: a saturated orange sky, a thick dust cloud, the handler pushed to the left edge, and water jars where the prompt asked for saddlebags and rugs.
Where the handler and the line ended up
The prompt puts the handler near the middle of a line of twelve. Muse Image and Wan 2.7 Image walk him mid-line. Muse Image draws a much longer caravan, about 19 camels, while Wan 2.7 Image has about a dozen. Nano Banana 2 put him at the front end of the line, FLUX.2 [flex] pushed him to the right, and Seedream 4.5 and Wan 2.6 put him at the left. Recraft V4.1 Pro and FLUX.2 [max] show a much shorter line.
Framing: from the dune crest, not at ground level
Only a few pictures look down on the caravan as asked. Muse Image, FIBO 2.0 and Cosmos 3 Super all shoot from above. FIBO 2.0 has no visible sun at all, so the golden-hour backlight is missing. HunyuanImage 3.0 Instruct is the opposite of an extreme wide shot: the handler and camels fill the frame at close range, so the sense of scale is lost.
Light and grade: amber and cobalt, or something else
Nano Banana Pro, Qwen Image 3.0 Pro and Muse Image hold the warm sun against a cool blue sky well. Grok Imagine Image pushes the sky to a deep navy that reads closer to dusk. Luma Uni-1 and Luma Uni-1 Max have a film-like softness, but the sun is washed out and the amber is weak. Seedream 5.0 Lite goes for a heavy, glossy sunset.
Details that give it away
Seedream 4.5 paints the lower legs of several camels solid black, which looks pasted on against the sunlit sand. Nano Banana Pro and the two Qwen models make the nearest camels huge and crowd the frame, which is dramatic but not the wide view requested.
Head to heads featuring these models
Two of the models above on the same film scenes, judged pair by pair.
- Nano Banana Pro vs Nano Banana 2
- FLUX.2 [pro] vs Nano Banana Pro
- Seedream 4.5 vs Nano Banana Pro
- Seedream 5.0 Lite vs Nano Banana 2
- Seedream 4.5 vs Seedream 5.0 Lite
- Qwen Image 3.0 vs FLUX.2 [pro]
- FLUX.2 [pro] vs FLUX.2 [max]
- FLUX.2 [pro] vs Seedream 4.5
- Grok Imagine Image vs Nano Banana 2
- HunyuanImage 3.0 Instruct vs Nano Banana Pro
Other scenes
The same 25 models on another film shot.
- Best AI image generator for cinematic portraits
- Best AI image generator for film noir
- Best AI image generator for thriller scenes
- Best AI image generator for drama scenes
- Best AI image generator for neon night scenes
- Best AI image generator for westerns
- Best AI image generator for action scenes
- Best AI image generator for fast motion
- Best AI image generator for storms and weather
Join the waitlist
Sign in once. If you don't have an invite yet, that puts you on the waitlist, and we'll email you when your invite is ready. For people 18 or over.
Join the waitlistSold under more than one name, shown once: Nano Banana 2 (Nano Banana 2, Gemini 3.1 Flash Image Preview); Nano Banana Pro (Nano Banana Pro, Gemini 3 Pro Image Preview).