The AI Model Leaderboards.
Top 10 models for every modality — image, video, speech, and music. Ranked by performance benchmarks, with pricing and access detail where available. Updated weekly.
Text-to-Image
Models that generate images from text prompts. Ranked by Artificial Analysis Image Arena ELO and community quality benchmarks.
| # | Model | Arena ELO | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
GPT Image 2 (high)
OpenAI
|
1339 | ~12s/img | ChatGPT integrated · arena leader | $0.05 / image | API |
| 2 |
MAI-Image-2.5
Microsoft AI
|
1273 | ~15s/img | Microsoft Foundry · new June 2 release | $0.04 / image | API |
| 3 |
HiDream-O1-Image-1.5
HiDream
|
1264 | ~14s/img | O1-line reasoning · text rendering specialist | $0.05 / image | API |
| 4 |
GPT Image 1.5 (high)
OpenAI
|
1262 | ~10s/img | Previous-gen ChatGPT image, still strong | $0.04 / image | API |
| 5 |
Nano Banana 2 (Gemini 3.1 Flash Image)
Google
|
1254 | ~8s/img | Photoreal · Gemini-integrated | $0.04 / image | API |
| 6 |
Reve 2.0
Reve
|
1218 | ~12s/img | Native 4K · June 3 release | $0.06 / image | API |
| 7 |
Midjourney v8
Midjourney
|
1212 | ~30s/img | Stylized leader · v8 ultra mode | Subscription · $10–60/mo | Subscription |
| 8 |
Ideogram 3.5
Ideogram
|
1198 | ~14s/img | Best-in-class text rendering in images | $0.08 / image · API tier | API |
| 9 |
Adobe Firefly Image 5
Adobe
|
1191 | ~12s/img | Commercial-safe training · CC integrated | Subscription · Adobe CC | Subscription |
| 10 |
Stable Diffusion 3.5 Large
Stability AI
|
1184 | ~6s/img on A100 | 8B params · open weights · run anywhere | Open source | Open Source |
Image Generation & Editing
Models that edit existing images via instruction, inpainting, or guided generation. Ranked by edit fidelity, instruction adherence, and quality preservation.
| # | Model | Edit Arena ELO | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
Flux 1.2 Fill Ultra
Black Forest Labs
|
1281 | ~10s/edit | Best instruction-following on edit prompts | $0.05 / edit | API |
| 2 |
GPT-Image 2 Edit
OpenAI
|
1262 | ~14s/edit | Conversational editing · multi-turn refinement | $0.05 / edit | API |
| 3 |
Imagen 4 Edit
Google
|
1248 | ~9s/edit | Mask-guided + instruction-guided edits | $0.05 / edit | API |
| 4 |
Adobe Firefly Edit 4
Adobe
|
1241 | ~11s/edit | Generative Fill in Photoshop · commercial-safe | Subscription · Adobe CC | Subscription |
| 5 |
Recraft Edit V3
Recraft
|
1227 | ~16s/edit | Brand-consistent edits · style locks | $0.05 / edit | API |
| 6 |
Ideogram Magic Fill 3.0
Ideogram
|
1213 | ~12s/edit | Text-preserving inpainting | $0.08 / edit | API |
| 7 |
Stable Diffusion 3.5 ControlNet
Stability AI
|
1206 | ~5s/edit on A100 | Full open ecosystem · LoRAs · masks | Open source | Open Source |
| 8 |
Midjourney Editor v7
Midjourney
|
1197 | ~25s/edit | Variations, vary region, zoom out | Subscription · $10–60/mo | Subscription |
| 9 |
FLUX.1 Kontext Pro
Black Forest Labs
|
1188 | ~7s/edit | Context-aware multi-image editing | $0.04 / edit | API |
| 10 |
Leonardo Canvas
Leonardo.ai
|
1172 | ~14s/edit | Inpaint, outpaint, sketch-to-image | Subscription · $10–48/mo | Subscription |
Text-to-Video
Models that generate video clips from text prompts. Ranked by motion quality, prompt adherence, and visual fidelity at native resolution.
| # | Model | Video Arena ELO | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
Dreamina Seedance 2.0 720p
ByteDance Seed
|
1219 | ~60s/clip | Up to 10s · 720p · arena leader | $9.07 / min | API |
| 2 |
HappyHorse-1.1
Alibaba-ATH
|
1151 | ~70s/clip | Up to 10s · 1080p · fast-rising | $9.90 / min | API |
| 3 |
HappyHorse-1.0
Alibaba-ATH
|
1123 | ~80s/clip | Up to 10s · 1080p | $13.20 / min | API |
| 4 |
SkyReels V4
Skywork AI
|
1106 | ~90s/clip | Up to 10s · 1080p · narrative-friendly | $21.00 / min | API |
| 5 |
Kling 3.0 1080p (Pro)
KlingAI
|
1104 | ~80s/clip | Up to 10s · 1080p · Pro tier | $20.16 / min | API |
| 6 |
Kling 3.0 Omni 1080p (Pro)
KlingAI
|
1098 | ~80s/clip | Up to 10s · 1080p · multimodal Omni | $16.80 / min | API |
| 7 |
Kling 3.0 720p (Standard)
KlingAI
|
1097 | ~70s/clip | Up to 10s · 720p · standard tier | $15.12 / min | API |
| 8 |
Kling 3.0 Omni 720p (Standard)
KlingAI
|
1094 | ~70s/clip | Up to 10s · 720p · multimodal Omni | $13.44 / min | API |
| 9 |
Veo 3.1
Google
|
1094 | ~90s/clip | Up to 8s · 1080p · synchronized audio | $24.00 / min | API |
| 10 |
Wan 2.7
Alibaba
|
1094 | ~80s/clip | Up to 10s · 1080p · multilingual prompts | $16.90 / min | API |
Image-to-Video
Models that animate or extend a still image into video. Ranked by motion realism, subject preservation, and edit-prompt adherence.
| # | Model | Video Arena ELO (I2V) | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
Dreamina Seedance 2.0 720p
ByteDance Seed
|
1194 | ~60s/clip | Up to 10s · 720p · I2V leader | $9.07 / min | API |
| 2 |
HappyHorse-1.1
Alibaba-ATH
|
1117 | ~70s/clip | Up to 10s · 1080p · fast-rising | $9.90 / min | API |
| 3 |
grok-imagine-video-1.5-preview
xAI
|
1110 | ~60s/clip | Up to 10s · 1080p · preview | $8.40 / min | API |
| 4 |
Wan 2.7
Alibaba
|
1090 | ~80s/clip | Up to 10s · 1080p · multilingual prompts | $16.90 / min | API |
| 5 |
HappyHorse-1.0
Alibaba-ATH
|
1089 | ~80s/clip | Up to 10s · 1080p | $13.20 / min | API |
| 6 |
Veo 3.1
Google
|
1087 | ~90s/clip | Up to 8s · 1080p · synchronized audio | $24.00 / min | API |
| 7 |
SkyReels V4
Skywork AI
|
1082 | ~90s/clip | Up to 10s · 1080p · narrative-friendly | $21.00 / min | API |
| 8 |
grok-imagine-video
xAI
|
1081 | ~55s/clip | Up to 10s · 1080p · cheapest I2V tier | $4.20 / min | API |
| 9 |
PixVerse V6
PixVerse
|
1075 | ~50s/clip | Up to 8s · 1080p | $6.90 / min | API |
| 10 |
Veo 3.1 Fast
Google
|
1075 | ~45s/clip | Up to 8s · 1080p · speed-optimized | $9.00 / min | API |
Speech-to-Text
Models that transcribe spoken audio into text. Ranked by Word Error Rate (lower is better) on standard benchmarks (LibriSpeech, FLEURS, AMI).
| # | Model | WER (lower = better) | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
Fun-Realtime-ASR-preview
Alibaba (Fun)
|
1.7% | realtime | AA-WER 1.7% · preview | Preview · contact | Preview |
| 2 |
Scribe v2
ElevenLabs
|
2.2% | realtime | AA-WER 2.2% · production | $3.67 / 1k min | API |
| 3 |
MAI-Transcribe-1.5
Microsoft Azure
|
2.4% | realtime | AA-WER 2.4% · Azure-native | $6.00 / 1k min | API |
| 4 |
Smallest AI Pulse Pro
Smallest.ai
|
2.4% | realtime | AA-WER 2.4% · low-cost tier | $3.50 / 1k min | API |
| 5 |
MAI-Transcribe-1
Microsoft Azure
|
2.6% | realtime | AA-WER 2.6% · prior Azure gen | $6.00 / 1k min | API |
| 6 |
Voxtral Small
Mistral
|
2.8% | realtime | AA-WER 2.8% | $4.00 / 1k min | API |
| 7 |
Gemini 3.1 Pro (High)
Google
|
2.8% | ~realtime | AA-WER 2.8% · Pro reasoning | $18.15 / 1k min | API |
| 8 |
Gemini 3 Flash (High)
Google
|
2.9% | realtime | AA-WER 2.9% · Flash tier | $13.70 / 1k min | API |
| 9 |
Gemini 2.5 Pro
Google
|
2.9% | ~realtime | AA-WER 2.9% · prior Pro gen | $11.39 / 1k min | API |
| 10 |
Solaria-3
Gladia
|
3.2% | realtime | AA-WER 3.2% · multilingual | $10.16 / 1k min | API |
Text-to-Speech
Models that generate spoken audio from text. Ranked by TTS Arena ELO and voice quality / naturalness benchmarks.
| # | Model | TTS Arena ELO | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
Sonic 3.5
Cartesia
|
1218 | ~realtime | Quality ELO 1218 · arena leader | $0.03 / 1k chars | API |
| 2 |
Gemini 3.1 Flash TTS
Google
|
1216 | ~realtime | Quality ELO 1216 · Flash-tier price | $0.075 / 1k chars | API |
| 3 |
Fun-Realtime-TTS
Alibaba (Fun)
|
1210 | realtime | Quality ELO 1210 · preview | Preview · contact | Preview |
| 4 |
Realtime TTS-2 (Research Preview)
OpenAI
|
1202 | realtime | Quality ELO 1202 · research preview | Preview · OpenAI tier | Preview |
| 5 |
xAI Text to Speech
xAI
|
1197 | ~realtime | Quality ELO 1197 | $0.06 / 1k chars | API |
| 6 |
ElevenLabs v3 Multilingual
ElevenLabs
|
1188 | ~realtime | 32 languages · vocal-realism leader | Subscription · $5–330/mo | Subscription |
| 7 |
PlayHT 3.0 Conversational
PlayHT
|
1172 | ~realtime | Conversational tuning · streaming | $0.06 / 1k chars · API tier | API |
| 8 |
Azure Neural Voice Pro
Microsoft Azure
|
1157 | ~realtime | Enterprise-grade · 140+ languages | $0.024 / 1k chars (Neural) | API |
| 9 |
OpenAI tts-1-hd
OpenAI
|
1148 | ~realtime | 6 voices · simple integration | $0.030 / 1k chars | API |
| 10 |
Coqui XTTS-v2
Coqui AI
|
1132 | ~3x realtime on A100 | Open weights · voice cloning · 17 languages | Open source | Open Source |
Music Generation
Models that generate full musical compositions from text prompts. Ranked by composite community evaluations and producer review scores — no consensus benchmark exists.
| # | Model | Composite Score | Speed | Output / Detail | Pricing | Access |
|---|---|---|---|---|---|---|
| 1 |
Suno v5
Suno
|
1293 | ~25s/song | Quality leader · vocal realism + structure · ELO 1293 | Subscription · $10–30/mo | Subscription |
| 2 |
Suno v5.5
Suno
|
1278 | ~25s/song | Refined v5 · easiest "idea → release" workflow | Subscription · $10–30/mo | Subscription |
| 3 |
Udio v1.5
Udio
|
1232 | ~30s/song | Stems + section regeneration · closest to AI DAW | Subscription · $10–30/mo | Subscription |
| 4 |
ElevenLabs Music
ElevenLabs
|
1218 | ~25s/song | Licensed-data trained · 44.1kHz · commercial-safe | Subscription · $5–330/mo | Subscription |
| 5 |
Google Lyria 3
Google
|
1205 | ~30s/song | Vocal-capable (new Feb 2026) · API/platform path | API · Vertex AI pricing | API |
| 6 |
Stable Audio 2.5
Stability AI
|
1187 | ~20s/song | Sound design + instrumental beds · commercial use | $0.10 / song · subscription tiers | API |
| 7 |
AIVA 4.0
AIVA
|
1172 | ~45s/song | Classical/cinematic specialist · MIDI export | Subscription · $11–48/mo | Subscription |
| 8 |
Riffusion FUZZ
Riffusion
|
1164 | ~25s/song | Full song generation · stems export | Subscription · $9–25/mo | Subscription |
| 9 |
MusicGen Large v2
Meta
|
1145 | ~30s/song on A100 | 3.3B params · open weights · text + melody conditioned | Open source | Open Source |
| 10 |
YuE Music v1
M-A-P
|
1128 | ~90s/song on A100 | 7B params · open weights · full vocal songs | Open source | Open Source |
Frequently asked questions.
How are the leaderboards ranked?
Each leaderboard uses the most credible benchmark for that modality. Text-to-Image and Text-to-Speech use Artificial Analysis Arena ELO. Speech-to-Text uses Artificial Analysis WER (AA-AgentTalk + VoxPopuli-Cleaned + Earnings22-Cleaned). Video (T2V + I2V) uses Artificial Analysis Video Arena. Music Generation uses a composite of Suno Arena ELO and community evaluations — methodology is credited on every section.
What's the best text-to-image model in 2026?
OpenAI GPT Image 2 leads the AA Image Arena at ELO 1339 as of June 2026, followed by Microsoft MAI-Image-2.5 (1273), HiDream-O1-Image-1.5 (1264), and Nano Banana 2 / Gemini 3.1 Flash Image (1254). For native 4K output, Reve 2.0 is the new June release worth tracking. Ideogram 3.5 remains the text-rendering specialist. For brand-safe commercial use, Adobe Firefly Image 5 leads. Best open-source: Stable Diffusion 3.5 Large.
What's the best text-to-video model in 2026?
ByteDance Dreamina Seedance 2.0 leads the AA Video Arena at ELO 1219 as of June 2026, with Alibaba HappyHorse-1.1 (1151) and HappyHorse-1.0 (1123) close behind. Skywork SkyReels V4 (1106) and the Kling 3.0 family (1094–1104) round out the top tier. Google Veo 3.1 (1094) remains the differentiator for synchronized audio. Note: OpenAI Sora 2 is deprecated (app retired April 26; API sunsets Sep 24, 2026) — not a current top pick.
What's the fastest speech-to-text model?
On the Artificial Analysis STT leaderboard (AA-AgentTalk + VoxPopuli + Earnings22), Alibaba's Fun-Realtime-ASR-preview leads accuracy at AA-WER 1.7% (preview). ElevenLabs Scribe v2 leads production-ready accuracy at 2.2% and $3.67/1k minutes. Microsoft MAI-Transcribe-1.5 follows at 2.4% / $6.00/1k min. All three run at realtime or near-realtime.
Why do some models show "Open Source" instead of a price?
Open-source models like Stable Diffusion 3.5, Whisper v4, MusicGen, and HunyuanVideo are released under permissive licenses and run on your own infrastructure. The "Pricing" cell shows "Open source" instead of a per-call price because the cost is your compute (GPU hours, electricity) rather than a vendor fee. This is performance ranking, not cost ranking — open-source models earn their rank on the same benchmarks as commercial APIs.
How often is this data updated?
The dataset is reviewed weekly against published benchmarks and vendor releases. New flagship model launches are typically added within 7 days. Quality scores update as new arena evaluations are published — the per-section "verified" badge on each section shows the most recent review date.
Why isn't there a single Quality Score across all leaderboards?
Each modality has its own canonical evaluation methodology. Image generation uses pairwise arena rankings (ELO). Speech-to-text uses Word Error Rate (a percentage). Text-to-speech uses arena ELO with different judging criteria. Normalizing these to a single 0-100 score would obscure the actual evaluation signal. We preserve the native benchmark per category and credit the source — builders can trust the signal more than a derivative composite.
Where can I see text-LLM rankings?
Text large language models have their own dedicated page: the LLM Cost Calculator. It tracks 100 top text-LLMs by Artificial Analysis Intelligence Index alongside cost, speed, latency, and context window. Link in the page header.
How do you handle models with subscription-only pricing?
Some models (Midjourney, Suno, Udio, Adobe Firefly) are subscription-only with no per-output API pricing. We show "Subscription · $X–$Y/mo" with the range, plus link to the vendor's pricing page. For builders comparing per-output cost specifically, we recommend either getting an API-accessible alternative or budgeting at the subscription tier.
Will you add other categories (3D generation, agents, embeddings)?
Probably — based on user demand. Text-to-3D, agentic frameworks, and embeddings are tracked as planned categories. Subscribe to the API waitlist above and we'll notify you when they ship.
Missing a model or category?
We add new entries when they earn a leaderboard spot. Suggest a model or modality — we review submissions within 7 days.