AI-model rangliste
Every model you can run on Railwail right now, by category: quality from the public Arena leaderboard, the live price of a default call, the context window and run times measured on our own jobs. Scores only where the arena rates exactly the model we run.
Arena scores published 22. sep. 2026Run times from Railwail jobs, last 90 days
- Models you can run
- 274
- Labs
- 34
- Categories
- 9
Today’s record holders
Best models by category
Quality from the public Arena, prices from our pricing rules. Open a category to sort by quality, price or value.
Text & chat
16 of 30 rated- Best quality
- Gemini 3.1 ProArena 1.487
- Cheapest
- Granite Code 8B0,12 US$ / 1M tokens
- Best value
- Gemini 3.1 Pro8 pts/$
Image generation
22 of 66 rated- Best quality
- Qwen Image 3 ProArena 1.255
- Cheapest
- Stable Diffusion XL≈ 0,0015 US$/kørsel
- Best value
- Nano Banana 2 Lite3.828 pts/$
Image editing
7 of 7 rated- Best quality
- Grok Imagine ImageArena 1.329
- Cheapest
- Grok Imagine Image0,024 US$/billede
- Best value
- Grok Imagine Image3.679 pts/$
Video generation
13 of 27 rated- Best quality
- Google Veo 3.1Arena 1.362
- Cheapest
- LTX-Video (Lightricks)≈ 0,0229 US$/kørsel
- Best value
- Grok Imagine Video504 pts/$
Image to video
14 of 14 rated- Best quality
- Grok Imagine VideoArena 1.415
- Cheapest
- Seedance Lite0,0432 US$/s
- Best value
- Grok Imagine Video556 pts/$
9 models
Filter by labTap a lab to filter every board.
01
Language models
30Chat, reasoning and code models billed per token. Sorted by Arena score, highest first; models without a rating follow by price.
| Actions | ||||||
|---|---|---|---|---|---|---|
Claude Opus 4.7 Arena 1494 · 30,00 US$/1M out · 1 mio. ctx | 1494 | 6,00 US$/1M in6,00 US$ | 30,00 US$ | 1 mio. | · | Try |
Gemini 3.1 Pro Arena 1487 · 14,40 US$/1M out · 1 mio. ctx | 1487 | 2,40 US$/1M in2,40 US$ | 14,40 US$ | 1 mio. | · | Try |
GPT-5.5 Arena 1476 · 36,00 US$/1M out · 400 t ctx | 1476 | 6,00 US$/1M in6,00 US$ | 36,00 US$ | 400 t | · | Try |
Gemini 3 Flash Arena 1474 · 3,60 US$/1M out · 1 mio. ctx | 1474 | 0,60 US$/1M in0,60 US$ | 3,60 US$ | 1 mio. | · | Try |
Claude Opus 4.8 Arena 1473 · 30,00 US$/1M out · 1 mio. ctx · median 2,3 s | 1473 | 6,00 US$/1M in6,00 US$ | 30,00 US$ | 1 mio. | 2,3 s15 runs | Try |
Claude Sonnet 4.6 Arena 1473 · 18,00 US$/1M out · 1 mio. ctx | 1473 | 3,60 US$/1M in3,60 US$ | 18,00 US$ | 1 mio. | · | Try |
GPT-5.4 Arena 1466 · 18,00 US$/1M out · 1,1 mio. ctx | 1466 | 3,00 US$/1M in3,00 US$ | 18,00 US$ | 1,1 mio. | · | Try |
Gemini 2.5 Pro Arena 1446 · 12,00 US$/1M out · 1 mio. ctx | 1446 | 1,50 US$/1M in1,50 US$ | 12,00 US$ | 1 mio. | · | Try |
GPT-5.1 Arena 1439 · 12,00 US$/1M out · 400 t ctx | 1439 | 1,50 US$/1M in1,50 US$ | 12,00 US$ | 400 t | · | Try |
OpenAI o3 Arena 1432 · 9,60 US$/1M out · 200 t ctx | 1432 | 2,40 US$/1M in2,40 US$ | 9,60 US$ | 200 t | · | Try |
| Claude Haiku 4.5 Anthropic | 1,20 US$/1M in | |||||
| GPT-4.1 OpenAI | 2,40 US$/1M in | |||||
| OpenAI o4-mini OpenAI | 1,32 US$/1M in | |||||
| o3-mini OpenAI | 1,32 US$/1M in | |||||
| GPT-4o OpenAI | 3,00 US$/1M in | |||||
| GPT-4o Mini OpenAI | 0,18 US$/1M in | |||||
| Granite Code 8B IBM | 0,060 US$/1M in | |||||
| GPT-6 Luna OpenAI | 0,12 US$/1M in | |||||
| Granite Code 20B IBM | 0,12 US$/1M in | |||||
| GPT-5.4 Nano OpenAI | 0,24 US$/1M in | |||||
| GPT-5 Mini OpenAI | 0,30 US$/1M in | |||||
| DeepSeek V4 Flash DeepSeek | 0,36 US$/1M in | |||||
| DeepSeek V4.1 Flash DeepSeek | 0,36 US$/1M in | |||||
| GPT-5.4 Mini OpenAI | 0,90 US$/1M in | |||||
| DeepSeek V4 Pro DeepSeek | 1,584 US$/1M in | |||||
| Claude Sonnet 5 Anthropic | 2,40 US$/1M in | |||||
| GPT-6 Sol OpenAI | 2,40 US$/1M in | |||||
| Claude Opus 5.5 Anthropic | 4,80 US$/1M in | |||||
| Claude Fable 5.1 Anthropic | 12,00 US$/1M in | |||||
| GPT-6 Astra OpenAI | 12,00 US$/1M in |
10 of 30 · Median run “·” = Fewer than 5 completed runs in 90 days
Arena scores: Text Arena leaderboard by LMArena, published 13. sep. 2026, CC BY 4.0.Ranked by quality, price and value →
A usage column appears once 5 models have runs from 3 or more different accounts in 30 days.
02
Image generation & editing
83/134Text-to-image, editing, upscaling and background removal. Sorted by Arena score, highest first; models without a rating follow by price.
| Actions | |||
|---|---|---|---|
Qwen Image 3 Pro Arena 1255 | 1255 | 0,048 US$/billede | Try |
Nano Banana 2 Lite Arena 1250 | 1250 | 0,0408 US$/billede | Try |
Ideogram V4 Quality Arena 1204 | 1204 | 0,12 US$/billede | Try |
Grok Imagine Image Arena 1171 | 1171 | 0,024 US$/billede | Try |
Recraft V4.1 Utility Pro Arena 1169 | 1169 | 0,30 US$/billede | Try |
Hunyuan Image 3 Arena 1151 | 1151 | 0,096 US$/billede | Try |
Nano Banana Arena 1150 | 1150 | 0,0468 US$/billede | Try |
Recraft V4.1 Pro Arena 1131 | 1131 | 0,30 US$/billede | Try |
Google Imagen 4 Arena 1129 | 1129 | 0,048 US$/billede | Try |
Qwen Image 2512 Arena 1125 | 1125 | 0,024 US$/billede | Try |
| Recraft V4 Recraft | 0,048 US$/billede | ||
| Flux Kontext Max Black Forest Labs | 0,096 US$/billede | ||
| Flux Kontext Pro Black Forest Labs | 0,048 US$/billede | ||
| Qwen Image Alibaba (Qwen) | 0,030 US$/billede | ||
| Ideogram v3 Quality Ideogram | 0,108 US$/billede | ||
| Photon Luma AI | 0,036 US$/billede | ||
| P Image Pruna AI | 0,0060 US$/billede | ||
| Gen4 Image Runway | 0,096 US$/billede | ||
| Recraft V3 Recraft | 0,048 US$/billede | ||
| FLUX 1.1 Pro Black Forest Labs | 0,048 US$/billede | ||
| Ideogram v2 Ideogram | 0,096 US$/billede | ||
| Stable Diffusion 3.5 Large Stability AI | 0,078 US$/billede | ||
| 851-Labs Background Remover Community | ≈ 0,00050 US$/kørsel | ||
| Remove Background (lucataco) Community | ≈ 0,00060 US$/kørsel | ||
| Remove Object (LaMa) Community | ≈ 0,00090 US$/kørsel | ||
| DDColor Community | ≈ 0,0012 US$/kørsel | ||
| Html To Image Community | 0,0012 US$/billede | ||
| Stable Diffusion XL Stability AI | ≈ 0,0015 US$/kørsel | ||
| Vectorizer (VTracer) Community | ≈ 0,0022 US$/kørsel | ||
| Real-ESRGAN 4x Community | 0,0024 US$/billede | ||
| SDXL Inpainting Community | ≈ 0,0026 US$/kørsel | ||
| GFPGAN v1.4 Tencent ARC | ≈ 0,0027 US$/kørsel | ||
| BiRefNet Background Removal Community | ≈ 0,0036 US$/kørsel | ||
| FLUX.1 [schnell] Black Forest Labs | 0,0036 US$/billede | ||
| FLUX.1 Redux Black Forest Labs | 0,0036 US$/billede | ||
| Consistent Character Community | ≈ 0,0038 US$/kørsel | ||
| InstructPix2Pix Community | ≈ 0,0039 US$/kørsel | ||
| CodeFormer Community | ≈ 0,0041 US$/kørsel | ||
| Rembg Community | ≈ 0,0047 US$/kørsel | ||
| Real-ESRGAN Anime 4x Community | ≈ 0,0051 US$/kørsel | ||
| Sticker Maker Community | ≈ 0,0055 US$/kørsel | ||
| Flux Fast Community | 0,0060 US$/billede | ||
| Hidream L1 Fast Community | 0,0060 US$/billede | ||
| PhotoMaker Tencent ARC | ≈ 0,0080 US$/kørsel | ||
| Swin2SR Community | ≈ 0,0088 US$/kørsel | ||
| Icons (SDXL Flat Pop) Community | ≈ 0,0094 US$/kørsel | ||
| Bringing Old Photos Back to Life Microsoft | ≈ 0,010 US$/kørsel | ||
| Face to Many Community | ≈ 0,0105 US$/kørsel | ||
| Image 01 MiniMax | 0,012 US$/billede | ||
| Photon Flash Luma AI | 0,012 US$/billede | ||
| Recraft Vectorize Recraft | 0,012 US$/billede | ||
| ControlNet Depth Community | ≈ 0,0144 US$/kørsel | ||
| SDXL Emoji Community | ≈ 0,0156 US$/kørsel | ||
| Janus Pro 7B DeepSeek | ≈ 0,0169 US$/kørsel | ||
| FLUX PuLID Community | ≈ 0,018 US$/kørsel | ||
| Lucid Origin Community | 0,0201 US$/billede | ||
| BRIA Remove Background Bria AI | 0,0216 US$/billede | ||
| Clarity Upscaler Community | ≈ 0,024 US$/kørsel | ||
| Hunyuan Image 2.1 Tencent | 0,024 US$/billede | ||
| Wan 2.2 Image Community | 0,024 US$/billede | ||
| Recraft 20B Recraft | 0,0264 US$/billede | ||
| IDM-VTON (Virtual Try-On) Community | ≈ 0,0277 US$/kørsel | ||
| IP-Adapter FaceID Plus v2 Community | ≈ 0,0277 US$/kørsel | ||
| Face to Sticker Community | ≈ 0,0289 US$/kørsel | ||
| Flux Krea Dev Black Forest Labs | 0,030 US$/billede | ||
| FLUX.1 [dev] Black Forest Labs | 0,030 US$/billede | ||
| FLUX.1 Canny [dev] Black Forest Labs | 0,030 US$/billede | ||
| FLUX.1 Depth [dev] Black Forest Labs | 0,030 US$/billede | ||
| FLUX.1 Kontext [dev] Black Forest Labs | 0,030 US$/billede | ||
| Ideogram V2A Turbo Ideogram | 0,030 US$/billede | ||
| Ideogram v3 Turbo Ideogram | 0,036 US$/billede | ||
| Qwen Image 3 Alibaba (Qwen) | 0,036 US$/billede | ||
| Qwen-Image-Edit Alibaba (Qwen) | 0,036 US$/billede | ||
| Seedream 4 ByteDance | 0,036 US$/billede | ||
| Wan 2.7 Image Alibaba (Wan) | 0,036 US$/billede | ||
| Wan 2.7 Image Pro Alibaba (Wan) | 0,036 US$/billede | ||
| LogoAI (SDXL Logo Generator) Community | ≈ 0,0372 US$/kørsel | ||
| Flux Dev Black Forest Labs | 0,0384 US$/billede | ||
| Qwen Image 2 Alibaba (Qwen) | 0,042 US$/billede | ||
| Seedream 5 Lite ByteDance | 0,042 US$/billede | ||
| Stable Diffusion 3 Stability AI | 0,042 US$/billede | ||
| Stable Diffusion 3.5 Medium Stability AI | 0,042 US$/billede | ||
| TRELLIS (3D) Community | ≈ 0,042 US$/kørsel | ||
| ESRGAN Classic Community | ≈ 0,0468 US$/kørsel | ||
| Fibo Bria AI | 0,048 US$/billede | ||
| FLUX.1 Fill [dev] Black Forest Labs | 0,048 US$/billede | ||
| Grok Imagine Image 2 xAI | 0,048 US$/billede | ||
| Ideogram V2A Ideogram | 0,048 US$/billede | ||
| Image 3.2 Bria AI | 0,048 US$/billede | ||
| Professional Headshot (FLUX Kontext) Black Forest Labs | 0,048 US$/billede | ||
| Recraft V4.1 Recraft | 0,048 US$/billede | ||
| Recraft V4.1 Utility Recraft | 0,048 US$/billede | ||
| Seedream 4.5 ByteDance | 0,048 US$/billede | ||
| Stable Diffusion 3.5 Large Turbo Stability AI | 0,048 US$/billede | ||
| Recraft 20B SVG Recraft | 0,0528 US$/billede | ||
| Magnific-Style Upscaler Community | ≈ 0,0589 US$/kørsel | ||
| FLUX.1 Canny Black Forest Labs | 0,060 US$/billede | ||
| FLUX.1 Depth Black Forest Labs | 0,060 US$/billede | ||
| FLUX.1 Fill Black Forest Labs | 0,060 US$/billede | ||
| Ideogram V2 Turbo Ideogram | 0,060 US$/billede | ||
| AuraFlow v0.3 Community | ≈ 0,0648 US$/kørsel | ||
| Flux Pro Black Forest Labs | 0,066 US$/billede | ||
| Ad Inpaint (Product Photo) Community | ≈ 0,0672 US$/kørsel | ||
| ControlNet Canny Community | ≈ 0,0697 US$/kørsel | ||
| Point-E Community | ≈ 0,0708 US$/kørsel | ||
| Flux 1.1 Pro Ultra Black Forest Labs | 0,072 US$/billede | ||
| Ideogram V3 Balanced Ideogram | 0,072 US$/billede | ||
| Ideogram V4 Balanced Ideogram | 0,072 US$/billede | ||
| Playground v2.5 (1024px Aesthetic) Playground AI | ≈ 0,0732 US$/kørsel | ||
| IC-Light (Product Relighting) Community | ≈ 0,0781 US$/kørsel | ||
| Nano Banana 2 Google DeepMind | 0,0804 US$/billede | ||
| InstantID Community | ≈ 0,084 US$/kørsel | ||
| Rd Animation Community | 0,084 US$/billede | ||
| Kuaishou Kolors Community | ≈ 0,0889 US$/kørsel | ||
| FLUX.1-dev Inpainting Community | ≈ 0,090 US$/kørsel | ||
| Qwen Image 2 Pro Alibaba (Qwen) | 0,090 US$/billede | ||
| Cartoonify Community | ≈ 0,096 US$/kørsel | ||
| Recraft v3 SVG Recraft | 0,096 US$/billede | ||
| Recraft V4 SVG Recraft | 0,096 US$/billede | ||
| Shap-E (OpenAI) Community | ≈ 0,1116 US$/kørsel | ||
| Gpt Image 2 OpenAI | 0,1536 US$/billede | ||
| Hunyuan3D 2.0 Tencent | ≈ 0,156 US$/kørsel | ||
| Gpt Image 1.5 OpenAI | 0,1632 US$/billede | ||
| Nano Banana Pro Google DeepMind | 0,18 US$/billede | ||
| OOTDiffusion (Try-On) Community | ≈ 0,1801 US$/kørsel | ||
| Hunyuan3D 2.1 Community | ≈ 0,264 US$/kørsel | ||
| CCSR (Content-Consistent SR) Community | ≈ 0,2761 US$/kørsel | ||
| Gpt Image 2.5 Flare OpenAI | 0,30 US$/billede | ||
| Gpt Image 2.5 Sunburst OpenAI | 0,30 US$/billede | ||
| Recraft V4 Pro Recraft | 0,30 US$/billede | ||
| SUPIR Upscaler Community | ≈ 0,30 US$/kørsel | ||
| SUPIR Community | ≈ 0,4921 US$/kørsel | ||
| StarVector 8B (image-to-SVG) Community | 0,00117 US$/GPU-s | ||
| DreamGaussian Community | 0,00168 US$/GPU-s |
10 of 83
Arena scores: Text-to-Image Arena leaderboard by LMArena, published 22. sep. 2026, CC BY 4.0.Ranked by quality, price and value →
03
Video generation
31/46Text-to-video and image-to-video. Sorted by Arena score, highest first; models without a rating follow by price.
| Actions | ||||
|---|---|---|---|---|
Google Veo 3.1 Arena 1362 | 1362 | 0,48 US$/s1,92 US$ for 4 s | · | Try |
Google Veo 3.1 Fast Arena 1358 | 1358 | 0,18 US$/s0,72 US$ for 4 s | · | Try |
Grok Imagine Video Arena 1342 · median 48 s | 1342 | 0,060 US$/s0,30 US$ for 5 s | 48 s31 runs | Try |
PixVerse v5.6 Arena 1240 | 1240 | 0,084 US$/s0,42 US$ for 5 s | · | Try |
Runway Gen 4.5 Arena 1225 | 1225 | 0,144 US$/s0,72 US$ for 5 s | · | Try |
Hailuo 2.3 Arena 1206 | 1206 | 0,336 US$/video | · | Try |
Seedance Pro Arena 1191 | 1191 | 0,18 US$/s0,90 US$ for 5 s | · | Try |
MiniMax Hailuo 02 Arena 1181 | 1181 | 0,324 US$/video | · | Try |
Google Veo 2 Arena 1164 | 1164 | 0,60 US$/s4,80 US$ for 8 s | · | Try |
Kling v2.1 Master Arena 1163 | 1163 | 0,336 US$/s1,68 US$ for 5 s | · | Try |
| Seedance Lite ByteDance | 0,0432 US$/s | |||
| Luma Ray-2 720p Luma AI | 0,216 US$/s | |||
| Mochi 1 Genmo | ≈ 0,5041 US$/kørsel | |||
| FILM Frame Interpolation Google Research | ≈ 0,0016 US$/kørsel | |||
| Wav2Lip Community | ≈ 0,0067 US$/kørsel | |||
| LTX-Video (Lightricks) Lightricks | ≈ 0,0229 US$/kørsel | |||
| SwinIR Video Community | ≈ 0,0277 US$/kørsel | |||
| Wan 2.2 5B Fast Alibaba (Wan) | 0,030 US$/video | |||
| RIFE Frame Interpolation Community | ≈ 0,0444 US$/kørsel | |||
| Wan 2.2 Image-to-Video Alibaba (Wan) | 0,060 US$/video | |||
| Wan 2.2 Text-to-Video Alibaba (Wan) | 0,060 US$/video | |||
| MuseTalk Community | ≈ 0,0624 US$/kørsel | |||
| ToonCrafter Community | ≈ 0,0852 US$/kørsel | |||
| LivePortrait Community | ≈ 0,0949 US$/kørsel | |||
| SadTalker Community | ≈ 0,1165 US$/kørsel | |||
| AnimateDiff Community | ≈ 0,1201 US$/kørsel | |||
| VideoCrafter Community | ≈ 0,1561 US$/kørsel | |||
| Kling v2.1 Kuaishou (Kling) | 0,060 US$/s | |||
| Runway Gen-4 Turbo Runway | 0,060 US$/s | |||
| Luma Ray Flash 2 Luma AI | 0,072 US$/s | |||
| Seedance 1 Pro Fast ByteDance | 0,072 US$/s | |||
| CogVideoX-5B (open) Zhipu AI | ≈ 0,4081 US$/kørsel | |||
| MagicAnimate Community | ≈ 0,4081 US$/kørsel | |||
| Kling V2.5 Turbo Pro Kuaishou (Kling) | 0,084 US$/s | |||
| EchoMimic Community | ≈ 0,48 US$/kørsel | |||
| Mochi 1 Community | ≈ 0,5041 US$/kørsel | |||
| CogVideoX-5B Community | ≈ 0,60 US$/kørsel | |||
| Minimax Video MiniMax | 0,60 US$/video | |||
| Wan 3 Alibaba (Qwen) | 0,12 US$/s | |||
| Google Veo 3 Fast Google DeepMind | 0,18 US$/s | |||
| Kling v3 Kuaishou (Kling) | 0,2688 US$/s | |||
| Kling v3 Omni Kuaishou (Kling) | 0,2688 US$/s | |||
| HunyuanVideo Tencent | ≈ 3,06 US$/kørsel | |||
| Google Veo 3 (Replicate) Google DeepMind | 0,48 US$/s | |||
| AnimateDiff Lightning Community | 0,00117 US$/GPU-s | |||
| V-Express Community | 0,00168 US$/GPU-s |
10 of 31 · Median run “·” = Fewer than 5 completed runs in 90 days
Arena scores: Text-to-Video Arena leaderboard by LMArena, published 22. sep. 2026, CC BY 4.0.Ranked by quality, price and value →
04
Text to speech
7/14Voices from text, including voice cloning. Sorted by the price of one default call, lowest first.
| Actions | ||
|---|---|---|
| 0,018 US$/1.000 tegn | Try | |
| 0,024 US$/1.000 tegn | Try | |
| 0,030 US$/1.000 tegn | Try | |
| 0,036 US$/1.000 tegn | Try | |
| Kokoro TTS 82M Community | ≈ 0,00030 US$/kørsel | |
| StyleTTS 2 Community | ≈ 0,00040 US$/kørsel | |
| Parler-TTS Community | ≈ 0,0024 US$/kørsel | |
| Spark TTS Community | ≈ 0,0035 US$/kørsel | |
| ≈ 0,0157 US$/kørsel | Try | |
| RVC Voice Conversion Community | ≈ 0,0505 US$/kørsel | |
| ≈ 0,0576 US$/kørsel | Try | |
| OpenVoice v2 Community | ≈ 0,0673 US$/kørsel | |
| Tortoise TTS Community | ≈ 0,0816 US$/kørsel | |
| ≈ 0,0972 US$/kørsel | Try |
05
Speech to text
5Transcription and speaker diarization. Sorted by the price of one default call, lowest first.
06
Vision & multimodal
30Open models that caption, read or segment images, billed by GPU time. Sorted by the price of one default call, lowest first.
| Actions | ||
|---|---|---|
| ≈ 0,00030 US$/kørsel | Try | |
| ≈ 0,00030 US$/kørsel | Try | |
| ≈ 0,00030 US$/kørsel | Try | |
| ≈ 0,00060 US$/kørsel | Try | |
| ≈ 0,0012 US$/kørsel | Try | |
| ≈ 0,0012 US$/kørsel | Try | |
| ≈ 0,0012 US$/kørsel | Try | |
| ≈ 0,0012 US$/kørsel | Try | |
| ≈ 0,0014 US$/kørsel | Try | |
| ≈ 0,0015 US$/kørsel | Try | |
| MiniCPM-V 2.6 Community | ≈ 0,0015 US$/kørsel | |
| Moondream2 Community | ≈ 0,0020 US$/kørsel | |
| Qwen2-VL 7B Instruct Community | ≈ 0,0024 US$/kørsel | |
| Llama 3.2 Vision 11B (Ollama) Community | ≈ 0,0039 US$/kørsel | |
| Depth Anything v2 Community | ≈ 0,0050 US$/kørsel | |
| Llama 3.2 Vision 90B Community | ≈ 0,0070 US$/kørsel | |
| DeepSeek-VL 7B DeepSeek | ≈ 0,0086 US$/kørsel | |
| GLPN Depth Community | ≈ 0,0094 US$/kørsel | |
| CogVLM2 19B Community | ≈ 0,0114 US$/kørsel | |
| Dots OCR Community | ≈ 0,0132 US$/kørsel | |
| SAM 2 (Segment Anything 2) Meta | ≈ 0,018 US$/kørsel | |
| olmOCR Community | ≈ 0,0205 US$/kørsel | |
| PaddleOCR v3 Community | ≈ 0,0216 US$/kørsel | |
| Mask2Former Meta | ≈ 0,0276 US$/kørsel | |
| Lotus-G Community | ≈ 0,0408 US$/kørsel | |
| CLIP Interrogator Community | ≈ 0,0457 US$/kørsel | |
| Molmo 7B Community | ≈ 0,0505 US$/kørsel | |
| Donut Document Community | ≈ 0,0648 US$/kørsel | |
| Marigold Community | ≈ 0,0936 US$/kørsel | |
| LLaVA 1.6 Vicuna 13B Community | ≈ 0,1021 US$/kørsel |
10 of 30
07
Open-weight text & code
10Open models on dedicated GPUs, billed by GPU time. Sorted by the price of one default call, lowest first.
08
Music & audio
3Music and sound generation. Sorted by the price of one default call, lowest first.
09
Embeddings
2Vectors for search and retrieval, billed per token. Sorted by input price per 1M tokens, lowest first.
| Actions | |||
|---|---|---|---|
OpenAI text-embedding-3-small 8,2 t ctx | 0,024 US$/1M in | 8,2 t | Try |
OpenAI text-embedding-3-large 8,2 t ctx | 0,156 US$/1M in | 8,2 t | Try |
10
How this leaderboard works
Every figure comes from our catalog, our pricing rules, our own job records or, for quality, the public Arena leaderboard. What we cannot source, we leave out.
- What is listed
- Only models you can run right now: active and with a verified price. Left out: 130 catalog entries without a price (they cannot run), 0 removed models and 5 duplicate entries of the same provider model. Legacy models stay listed and link their successor.
- Price
- What one call with the model’s default settings costs on Railwail, from the same pricing rules that bill your runs (1 credit = $0.01). Token models show the price per 1M input and output tokens. Prices marked ≈ are billed by GPU time: the figure is a typical run, the charge follows the real run time.
- Median run time
- Measured on completed Railwail jobs: the median of the newest up to 500 runs in the last 90 days, shown from 5 runs. It includes queueing at the provider. For language models it depends on the answer length, so the LLM board is never sorted by speed. Measured today: 3 models.
- Quality
- Arena scores from the public Arena leaderboard (formerly LMArena, CC BY 4.0), where people compare two anonymous models on the same prompt and vote. A model gets a score only when the arena entry is exactly the model and default setting we run (reviewed list, 56 models today). No score is estimated.
- Context window
- The maximum tokens per request as listed in our catalog. Not re-measured by us.
- Usage
- Runs per model are only shown once they say something: at least 5 models with runs from 3 or more different accounts in 30 days. Until then there is no popularity ranking.
- What we don’t show
- No votes of our own and no estimated scores: our own arena has 0 decided battles and 2 model votes, far too few for a rating. Models the public Arena has not rated in the setting we run show “No Arena rating”.
Source:arena.ai leaderboardDatasetCC BY 4.0
Found a wrong price or a missing model? Write to us
Questions about the leaderboard
Which language model is the cheapest on Railwail?
Right now Granite Code 8B by IBM at 0,060 US$/1M in (output 0,30 US$/1M out). The language model board lists all 30 language models by input price.
Which model has the largest context window?
GPT-5.4 by OpenAI with 1,1 mio. tokens per request, as listed in our catalog.
What is the lowest price per image?
FLUX.1 [schnell] by Black Forest Labs at 0,0036 US$/billede. Models billed by GPU time can cost less per run; their price, marked ≈, is a typical run and the charge follows the real run time.
Where do the quality scores come from?
From the public Arena leaderboard (formerly LMArena): people compare two anonymous models on the same prompt and vote for the better result; the votes give each model a score. We use the published leaderboards (text: 13. sep. 2026, image and video: 22. sep. 2026) under CC BY 4.0 and show a score only where the arena rates exactly the model and default setting that runs on Railwail.
How current are the prices?
They come from the same pricing rules that bill your runs and were last checked on 24. sep. 2026. The snapshot on this page is refreshed about once an hour.
How do I run one of these models?
Open the model and try it on its page, or create an account, get an API key and call it through the Railwail API at https://railwail.com/api/v1. Your balance is prepaid in USD; runs are billed by the same pricing rules as the prices shown here.
Keep exploring
One key, one balance in USD, every model on this page.
Get an API key