Timelines · leaderboards
Leaderboard history.
Rank and score over time for 24 boards. A new snapshot is saved whenever a board changes, so nothing is overwritten. Tracking since Oct 16, 2026.
Biggest moves
- Step Audio EditX (Mar 2026) moved from #25 to #41 on Open-weights TTS (AA subset) (2026-10-17 → 2026-10-23)
- StepAudio 2.5 TTS (Aug 2026) moved from #24 to #14 on Text-to-Speech: Artificial Analysis (2026-10-17 → 2026-10-23)
- Higgs Audio V3 TTS moved from #61 to #70 on Open-weights TTS (AA subset) (2026-10-17 → 2026-10-23)
- Kokoro 82M v1.0 moved from #49 to #57 on Open-weights TTS (AA subset) (2026-10-17 → 2026-10-23)
- Eleven v3 moved from #14 to #21 on Text-to-Speech: Artificial Analysis (2026-10-17 → 2026-10-23)
- Chatterbox moved from #72 to #78 on Open-weights TTS (AA subset) (2026-10-17 → 2026-10-23)
- v3 Conversational moved from #10 to #15 on Text-to-Speech: Artificial Analysis (2026-10-17 → 2026-10-23)
- Gemini 3.8 Flash-Lite TTS moved from #4 to #8 on Text-to-Speech: Artificial Analysis (2026-10-17 → 2026-10-23)
Coding and LLM boards
Aider polyglot
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2025-08-25 · aider.chat ↗
| # | Model | Score | Change | History |
|---|---|---|---|---|
| — | gpt-5 (high) | 88 | — | · |
| — | gpt-5 (medium) | 86.7 | — | · |
| — | o3-pro (high) | 84.9 | — | · |
| — | gemini-2.5-pro-preview-06-05 (32k think) | 83.1 | — | · |
| — | gpt-5 (low) | 81.3 | — | · |
| — | o3 (high) | 81.3 | — | · |
| — | grok-4 (high) | 79.6 | — | · |
| — | gemini-2.5-pro-preview-06-05 (default think) | 79.1 | — | · |
| — | o3 (high) + gpt-4.1 | 78.2 | — | · |
| — | o3 | 76.9 | — | · |
| — | Gemini 2.5 Pro Preview 05-06 | 76.9 | — | · |
| — | DeepSeek-V3.2-Exp (Reasoner) | 74.2 | — | · |
Android Bench 1.0 tasks (100 tasks)
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-08-11 · developer.android.com ↗
| # | Model | Score % (avg of 100 tasks) | Change | History |
|---|---|---|---|---|
| — | Claude Opus 5 | 91.8 | — | · |
| — | Claude Fable 5 | 90.9 | — | · |
| — | GPT 5.6 Sol | 90.8 | — | · |
| — | Kimi K3 | 90.2 | — | · |
| — | GPT 5.6 Luna | 87.6 | — | · |
| — | Qwen3.8 Max | 87 | — | · |
| — | GPT 5.6 Terra | 86.8 | — | · |
| — | GPT 5.5 | 80.2 | — | · |
| — | Claude Sonnet 5 | 76.2 | — | · |
| — | Gemini 3.6 Flash | 75.6 | — | · |
| — | Gemini 3.1 Pro Preview | 74.3 | — | · |
| — | GPT 5.4 | 74.1 | — | · |
| — | Claude Opus 4.8 | 72.4 | — | · |
| — | GLM 5.2 | 72.2 | — | · |
| — | Gemini 3.5 Flash | 71.1 | — | · |
Top 15 of 32 shown. Every snapshot keeps the full board.
Android Bench 2.0 (long-horizon, 30 tasks)
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-10-06 · developer.android.com ↗
| # | Model | Pass rate (%) | Change | History |
|---|---|---|---|---|
| — | Claude Opus 5.5 | 32.7 | — | · |
| — | GPT 6 Astra | 28 | — | · |
| — | Claude Fable 5.1 | 22.7 | — | · |
| — | GPT 6 Sol | 18 | — | · |
| — | Qwen3.8 Max | 14 | — | · |
| — | Kimi K3 | 12 | — | · |
| — | GPT 6 Luna | 10.7 | — | · |
| — | Gemini 3.8 Flash | 8 | — | · |
| — | Gemini 3.7 Flash | 7.3 | — | · |
| — | Claude Sonnet 5 | 6.7 | — | · |
Design Arena (All categories)
Tracking since Oct 16, 2026 · 4 snapshots · board dated 2026-10-16 → 2026-10-19 → 2026-10-23 → 2026-10-24 · designarena.ai ↗
- deepseek-v4-1-flash-1017 moved from #20 to #24 (2026-10-23 → 2026-10-24)
- claude-sonnet-5 moved from #38 to #35 (2026-10-23 → 2026-10-24)
- gemini-3.8-flash moved from #23 to #21 (2026-10-23 → 2026-10-24)
- gpt-6.1-sol moved from #7 to #6 (2026-10-23 → 2026-10-24)
- claude-sonnet-5-5 moved from #6 to #7 (2026-10-23 → 2026-10-24)
- colorado moved from #21 to #20 (2026-10-23 → 2026-10-24)
| # | Model | Elo (human preference) | Change | History |
|---|---|---|---|---|
| 1 | claude-opus-5-5 | 1389 | = | |
| 2 | gpt-6-astra | 1383 | = | |
| 3 | muse-wonder-1.0 | 1377 | = | |
| 4 | gpt-6-astra-max | 1371 | = | |
| 5 | kimi-k3 | 1359 | = | |
| 6 | gpt-6.1-sol | 1358 | ▲ 1 | |
| 7 | claude-sonnet-5-5 | 1354 | ▼ 1 | |
| 8 | muse-spark-1.3 | 1350 | = | |
| 9 | muse-spark-1.3-max | 1349 | = | |
| 10 | gpt-5.6-sol-xhigh | 1340 | = | |
| 11 | claude-fable-5-1 | 1334 | = | |
| 12 | gemini-4-argon | 1329 | = | |
| 13 | claude-opus-5 | 1322 | = | |
| 14 | deepseek-v4-1-flash | 1318 | = | |
| 15 | hampshire | 1315 | = |
Top 15 of 65 shown. Every snapshot keeps the full board.
Design Arena (Fullstack)
Tracking since Oct 16, 2026 · 3 snapshots · board dated 2026-10-16 → 2026-10-19 → 2026-10-23 · designarena.ai ↗
- No rank changes; scores may have moved slightly.
| # | Model | Elo (human preference) | Change | History |
|---|---|---|---|---|
| 1 | gpt-6-astra | 1357 | = | |
| 2 | claude-fable-5-1 | 1326 | = | |
| 3 | kimi-k3 | 1315 | = | |
| 4 | muse-spark-1.3-max | 1312 | = | |
| 5 | claude-opus-5 | 1306 | = | |
| 6 | qwen3.8-max | 1305 | = | |
| 7 | muse-spark-1.3 | 1295 | = | |
| 8 | claude-fable-5 | 1280 | = | |
| 9 | grogu | 1264 | = | |
| 10 | gpt-5.6-sol-xhigh | 1261 | = | |
| 11 | claude-sonnet-5 | 1249 | = | |
| 12 | yoda | 1249 | = | |
| 13 | claude-opus-4-8 | 1249 | = | |
| 14 | gemini-3.8-flash | 1244 | = | |
| 15 | glm-5.2 | 1241 | = |
Top 15 of 42 shown. Every snapshot keeps the full board.
LMArena WebDev Arena
Tracking since Oct 16, 2026 · 4 snapshots · board dated 2026-10-13 → 2026-10-18 → 2026-10-22 · lmarena.ai ↗
- qwen3.8-max-0902 moved from #10 to #12 (2026-10-22 → 2026-10-22)
- Gemini 3.8 Flash (High) moved from #34 to #36 (2026-10-22 → 2026-10-22)
- Claude Opus 5 (Max) moved from #8 to #9 (2026-10-22 → 2026-10-22)
- gpt-6-sol-max moved from #9 to #10 (2026-10-22 → 2026-10-22)
- qwen3.8-max moved from #12 to #13 (2026-10-22 → 2026-10-22)
- deepseek-v4.1-flash-20261017-max moved from #13 to #14 (2026-10-22 → 2026-10-22)
- Kimi K3.1 Pro (Max) joined at #8
| # | Model | Score | Change | History |
|---|---|---|---|---|
| 1 | claude-opus-5.5-max | 1807 | = | |
| 2 | gpt-6-astra-max | 1774 | = | |
| 3 | claude-sonnet-5.5-xhigh | 1769 | = | |
| 4 | gpt-6.1-sol-max | 1749 | = | |
| 5 | Claude Fable 5.1 (Max) | 1742 | = | |
| 6 | Muse Wonder Full (Max) | 1726 | = | |
| 7 | claude-sonnet-5.5-high | 1709 | = | |
| 8 | Kimi K3.1 Pro (Max) | 1692 | new | · |
| 9 | Claude Opus 5 (Max) | 1691 | ▼ 1 | |
| 10 | gpt-6-sol-max | 1681 | ▼ 1 | |
| 11 | Gemini 4 Argon (High) | 1674 | = | |
| 12 | qwen3.8-max-0902 | 1673 | ▼ 2 | |
| 13 | qwen3.8-max | 1672 | ▼ 1 | |
| 14 | deepseek-v4.1-flash-20261017-max | 1665 | ▼ 1 | |
| 15 | Claude Opus 5 (High) | 1659 | ▼ 1 |
Top 15 of 68 shown. Every snapshot keeps the full board.
SWE-Bench Pro (Scale, public set)
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-10-16 · scale.com ↗
| # | Model | Score | Change | History |
|---|---|---|---|---|
| 1 | Muse Spark 1.1 | 61.5 | — | · |
| 2 | gpt-5.4 (xHigh) | 59.1 | — | · |
| 3 | Muse Spark | 55 | — | · |
| 4 | claude-opus-4-6 (thinking) | 51.9 | — | · |
| 5 | gemini-3.1-pro (thinking) | 46.1 | — | · |
| 13 | qwen3-coder-480b-a35b | 38.7 | — | · |
| 14 | minimax-2.1 | 36.81 | — | · |
| 17 | kimi-k2-instruct | 27.67 | — | · |
| 18 | qwen3-235b-a22b | 21.41 | — | · |
| 19 | gpt-oss-120b | 16.2 | — | · |
| 20 | deepseek-v3p2 | 15.56 | — | · |
| 21 | gemma-3-27b-it | 11.38 | — | · |
| 22 | llama3-1-405b-instruct | 11.18 | — | · |
| 23 | glm-4.6 | 9.67 | — | · |
| 24 | llama4-maverick-17b-instruct | 5.24 | — | · |
Top 15 of 16 shown. Every snapshot keeps the full board.
SWE-bench Verified (Bash Only)
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-02-17 · swebench.com ↗
| # | Model | % Resolved (500 tasks) | Change | History |
|---|---|---|---|---|
| — | Claude 4.5 Opus (high) | 76.8 | — | · |
| — | Gemini 3 Flash (high) | 75.8 | — | · |
| — | MiniMax M2.5 (high) | 75.8 | — | · |
| — | Claude 4.6 Opus | 75.6 | — | · |
| — | GLM 5 (high) | 72.8 | — | · |
| — | GPT 5.2 (high) | 72.8 | — | · |
| — | GPT 5.2 Codex | 72.8 | — | · |
| — | Claude 4.5 Sonnet (high) | 71.4 | — | · |
| — | Kimi K2.5 (high) | 70.8 | — | · |
| — | DeepSeek V3.2 (high) | 70 | — | · |
| — | Gemini 3 Pro | 69.6 | — | · |
| — | Claude 4.5 Haiku (high) | 66.6 | — | · |
| — | GPT 5 mini | 56.2 | — | · |
SWE-rebench
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-07-01 · swe-rebench.com ↗
| # | Model | Score | Change | History |
|---|---|---|---|---|
| 1 | Fable 5 [high] | 64.5 | — | · |
| 2 | Grok 4.5 [high] | 63.8 | — | · |
| 3 | Opus 5 [high] | 63.4 | — | · |
| 4 | GLM-5.2 [high] | 62.9 | — | · |
| 5 | GPT-5.6 Sol [medium] | 62.3 | — | · |
| 6 | Junie Agent | 61.8 | — | · |
| 7 | Claude Code Agent | 60.4 | — | · |
| 8 | Codex Agent | 58 | — | · |
| 9 | Sonnet 5 [high] | 56.8 | — | · |
| 10 | Cursor Agent | 51.7 | — | · |
| 11 | MiniMax M3 | 47.2 | — | · |
| 12 | MiMo V2.5 Pro | 46.5 | — | · |
| 13 | GPT-5.6 Luna [medium] | 43.6 | — | · |
| 14 | DeepSeek-V4 Pro [high] | 40.2 | — | · |
| 15 | Qwen3.6-27B | 31.2 | — | · |
Top 15 of 17 shown. Every snapshot keeps the full board.
Terminal-Bench 4.0
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-09-03 · tbench.ai ↗
| # | Model | Resolution rate (%) | Change | History |
|---|---|---|---|---|
| — | Opus 5.5 (max) | 64.85 | — | · |
| — | Sonnet 5.5 (max) | 61.82 | — | · |
| — | GPT-6 Astra (max) | 58.18 | — | · |
| — | GPT-6.1 Sol (max) | 58.18 | — | · |
| — | Fable 5.1 (max) | 57.88 | — | · |
| — | GPT-6 Astra (xhigh) | 57.88 | — | · |
| — | GPT-6 Astra (high) | 57.88 | — | · |
| — | Fable 5.1 (xhigh) | 57.88 | — | · |
| — | Fable 5.1 (high) | 54.55 | — | · |
| — | GPT-6 Astra (medium) | 54.24 | — | · |
| — | Opus 5 (xhigh) | 53.94 | — | · |
| — | Fable 5.1 (medium) | 53.94 | — | · |
| — | Opus 5 (max) | 51.82 | — | · |
| — | GPT-6 Astra (low) | 50.61 | — | · |
| — | Opus 5 (high) | 50.3 | — | · |
Top 15 of 35 shown. Every snapshot keeps the full board.
Vals AI — LiveCodeBench
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-09-01 · vals.ai ↗
| # | Model | Score | Change | History |
|---|---|---|---|---|
| 1 | Claude Fable 5.1 | 90.52 | — | · |
| 2 | Claude Fable 5 | 89.78 | — | · |
| 3 | Gemini 3.8 Flash | 89.48 | — | · |
| 4 | Claude Opus 5 | 89.03 | — | · |
| 5 | Gemini 3.7 Flash | 88.65 | — | · |
| 7 | Grok 4.6 | 88.22 | — | · |
| 10 | Qwen 3.8 Max | 87.85 | — | · |
| 13 | DeepSeek V4 Pro 0813 | 87.53 | — | · |
| 15 | Grok 4.5 | 87.35 | — | · |
| 17 | DeepSeek V4 Flash 0731 | 87.26 | — | · |
| 18 | Kimi K3 | 87.19 | — | · |
| 29 | Muse Spark 1.1 | 85.91 | — | · |
| 39 | Grok 4.3 | 84.49 | — | · |
| 43 | Qwen 3.8 27B | 84 | — | · |
| 51 | GPT OSS 120B | 83.23 | — | · |
Top 15 of 25 shown. Every snapshot keeps the full board.
Vals AI — SWE-bench Verified
Tracking since Oct 16, 2026 · 1 snapshot · board dated 2026-09-01 · vals.ai ↗
| # | Model | % resolved (500 tasks) | Change | History |
|---|---|---|---|---|
| 1 | Claude Opus 5 | 97 | — | · |
| 2 | DeepSeek V4 Pro 0813 | 96.4 | — | · |
| 3 | GPT-5.6 Sol | 96.2 | — | · |
| 4 | Grok 4.6 | 95.6 | — | · |
| 5 | GPT-5.6 Terra | 95.4 | — | · |
| 6 | GLM 5.3 | 95.4 | — | · |
| 7 | Claude Fable 5 | 95 | — | · |
| 8 | Kimi K3 | 93.4 | — | · |
| 9 | GPT-5.6 Luna | 93 | — | · |
| 10 | GLM 5.3 Flash | 92 | — | · |
| 11 | DeepSeek V4 Flash 0731 | 88.8 | — | · |
| 12 | Claude Opus 4.8 | 88.6 | — | · |
| 13 | Grok 4.5 | 86.6 | — | · |
| 14 | Muse Spark 1.2 | 86.6 | — | · |
| 15 | Qwen 3.8 27B | 86 | — | · |
Top 15 of 30 shown. Every snapshot keeps the full board.
Vals AI Vibe Code Bench v1.1
Tracking since Oct 16, 2026 · 3 snapshots · board dated 2026-10-12 → 2026-10-18 → 2026-10-20 · vals.ai ↗
- MiniMax-M3 moved from #63 to #67 (2026-10-18 → 2026-10-20)
- GLM 5.3 Flash moved from #73 to #77 (2026-10-18 → 2026-10-20)
- Muse Spark moved from #83 to #87 (2026-10-18 → 2026-10-20)
- Grok 4.3 moved from #84 to #88 (2026-10-18 → 2026-10-20)
- Grok Build 0.1 moved from #92 to #96 (2026-10-18 → 2026-10-20)
- Mistral Medium 3.5 moved from #108 to #112 (2026-10-18 → 2026-10-20)
- Qwen 3.8 Max 0902 joined at #33
- MiniMax-M3.1-Flash joined at #39
| # | Model | Score | Change | History |
|---|---|---|---|---|
| 1 | Claude Sonnet 5.5 | 92.39 | = | |
| 2 | Gemini 4 Argon | 91.91 | = | |
| 3 | Claude Haiku 5.5 | 90.44 | = | |
| 4 | Claude Fable 5 | 90.35 | = | |
| 5 | Claude Opus 5.5 | 90.29 | = | |
| 6 | Claude Fable 5.1 | 90.26 | = | |
| 7 | GPT-6 Astra | 89.59 | = | |
| 8 | GPT-6.1 Sol | 88.93 | = | |
| 9 | Claude Opus 5 | 88.4 | = | |
| 10 | GPT-6 Sol | 87.82 | = | |
| 11 | DeepSeek V4.1 Flash 1017 | 87.73 | = | |
| 12 | Grok 4.7 | 86.17 | = | |
| 13 | Wonder Full | 86 | = | |
| 14 | Muse Spark 1.3 Max | 85.86 | = | |
| 15 | MiMo V2.6 Pro | 85.22 | = |
Top 15 of 44 shown. Every snapshot keeps the full board.
Video boards
Image-to-Video: Artificial Analysis
Tracking since Oct 19, 2026 · 2 snapshots · board dated 2026-10-17 → 2026-10-23 · artificialanalysis.ai ↗
- Seedance 2.5 moved from #2 to #3 (2026-10-17 → 2026-10-23)
- Gemini Omni Flash 1.1 moved from #3 to #4 (2026-10-17 → 2026-10-23)
- Wan 3.0 moved from #4 to #5 (2026-10-17 → 2026-10-23)
- MiniMax H3 Max moved from #5 to #6 (2026-10-17 → 2026-10-23)
- Seedance 2.0 moved from #6 to #7 (2026-10-17 → 2026-10-23)
- FLUX 3 moved from #7 to #8 (2026-10-17 → 2026-10-23)
- FLUX 3 Max (720p) joined at #2
- MAGI-2 Preview left the board (was #10)
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | MiniMax H3 | 1140 | = | |
| 2 | FLUX 3 Max (720p) | 1138 | new | · |
| 3 | Seedance 2.5 | 1138 | ▼ 1 | |
| 4 | Gemini Omni Flash 1.1 | 1132 | ▼ 1 | |
| 5 | Wan 3.0 | 1127 | ▼ 1 | |
| 6 | MiniMax H3 Max | 1125 | ▼ 1 | |
| 7 | Seedance 2.0 | 1121 | ▼ 1 | |
| 8 | FLUX 3 | 1086 | ▼ 1 | |
| 9 | Grok Imagine Video 1.5 | 1067 | ▼ 1 | |
| 10 | HappyHorse-1.1 | 1061 | ▼ 1 |
1 model no longer on this board (kept for history)
- MAGI-2 Preview: last seen 2026-10-17 at #10
Image-to-Video: DesignArena (transcribed from a post)
Tracking since Oct 19, 2026 · 1 snapshot · board dated 2026-10-16 · x.com ↗
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | HeyGen Video 1.0 | 1331 | — | · |
| 2 | MiniMax H3 Max | 1324 | — | · |
| 3 | P-Video-2-Pro (Speed) | 1324 | — | · |
| 4 | P-Video-2-Pro (Quality) | 1322 | — | · |
| 5 | P-Video-2-Pro | 1320 | — | · |
| 6 | Wan 3.0 | 1307 | — | · |
| 7 | MiniMax H3 | 1303 | — | · |
| 8 | Seedance 2.5 | 1287 | — | · |
| 9 | Grok Imagine Video 1.5 Preview | 1280 | — | · |
| 10 | Seedance 2.0 | 1266 | — | · |
| 11 | Seedance 2.0 Fast | 1260 | — | · |
| 12 | Happy Horse 1.1 | 1256 | — | · |
| 13 | FLUX 3 Video | 1251 | — | · |
| 14 | Grok Imagine | 1235 | — | · |
| 15 | Seedance 2.0 Mini | 1231 | — | · |
Top 15 of 20 shown. Every snapshot keeps the full board.
Image-to-Video: LMArena
Tracking since Oct 19, 2026 · 1 snapshot · board dated 2026-09-21 · lmarena.ai ↗
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | MiniMax H3 | 1495 | — | · |
| 2 | Gemini Omni 1.1 Flash | 1488 | — | · |
| 3 | Wan 3.0 | 1480 | — | · |
| 4 | Seedance 2.5 | 1477 | — | · |
| 5 | Seedance 2.0 | 1475 | — | · |
| 6 | gemini-omni-flash | 1465 | — | · |
| 7 | HiDream-O1-Video | 1456 | — | · |
| 8 | Grok Imagine Video 1.5 | 1456 | — | · |
| 9 | FLUX 3 Video | 1449 | — | · |
| 10 | HappyHorse 1.0 | 1442 | — | · |
Text-to-Video (with audio): Artificial Analysis
Tracking since Oct 19, 2026 · 2 snapshots · board dated 2026-10-17 → 2026-10-23 · artificialanalysis.ai ↗
- Dreamina Seedance 2.5 moved from #3 to #5 (2026-10-17 → 2026-10-23)
- MiniMax H3 Max moved from #5 to #7 (2026-10-17 → 2026-10-23)
- FLUX 3 moved from #6 to #8 (2026-10-17 → 2026-10-23)
- Seedance 2.0 moved from #7 to #9 (2026-10-17 → 2026-10-23)
- Gemini Omni Flash 1.1 moved from #8 to #10 (2026-10-17 → 2026-10-23)
- Agnes-Video-2.5 moved from #9 to #11 (2026-10-17 → 2026-10-23)
- HeyGen Video 1 joined at #3
- FLUX 3 Max (720p) joined at #6
- Wan2.7-260612 joined at #15
- Kling 3.0 Omni 1080p (Pro) joined at #17
- Kling 3.0 1080p (Pro) joined at #18
- Grok Imagine Video 1.5 Lite joined at #19
- Wan2.7 left the board (was #13)
- Kling 3.0 Omni 1080p Pro left the board (was #15)
- Kling 3.0 1080p Pro left the board (was #16)
- Grok Imagine 1.5 Lite left the board (was #17)
- Veo 3.1 left the board (was #19)
- Veo 3.1 Fast left the board (was #20)
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | Wan 3.0 | 1175 | = | |
| 2 | Utopai X | 1167 | = | |
| 3 | HeyGen Video 1 | 1160 | new | · |
| 4 | MiniMax H3 (768p) | 1158 | = | |
| 5 | Dreamina Seedance 2.5 | 1158 | ▼ 2 | |
| 6 | FLUX 3 Max (720p) | 1155 | new | · |
| 7 | MiniMax H3 Max | 1154 | ▼ 2 | |
| 8 | FLUX 3 | 1145 | ▼ 2 | |
| 9 | Seedance 2.0 | 1136 | ▼ 2 | |
| 10 | Gemini Omni Flash 1.1 | 1130 | ▼ 2 | |
| 11 | Agnes-Video-2.5 | 1066 | ▼ 2 | |
| 12 | Grok Imagine Video 1.5 | 1056 | ▼ 2 | |
| 13 | HappyHorse-1.1 | 1051 | ▼ 2 | |
| 14 | MAGI-2 Preview | 1042 | ▼ 2 | |
| 15 | Wan2.7-260612 | 1037 | new | · |
Top 15 of 20 shown. Every snapshot keeps the full board.
6 models no longer on this board (kept for history)
- Wan2.7: last seen 2026-10-17 at #13
- Kling 3.0 Omni 1080p Pro: last seen 2026-10-17 at #15
- Kling 3.0 1080p Pro: last seen 2026-10-17 at #16
- Grok Imagine 1.5 Lite: last seen 2026-10-17 at #17
- Veo 3.1: last seen 2026-10-17 at #19
- Veo 3.1 Fast: last seen 2026-10-17 at #20
Text-to-Video: LMArena
Tracking since Oct 19, 2026 · 1 snapshot · board dated 2026-09-21 · lmarena.ai ↗
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | Gemini Omni 1.1 Flash | 1516 | — | · |
| 2 | gemini-omni-flash | 1513 | — | · |
| 3 | FLUX 3 Video | 1493 | — | · |
| 4 | Grok Imagine Video 1.5 | 1492 | — | · |
| 5 | Seedance 2.0 | 1479 | — | · |
| 6 | Wan 3.0 | 1476 | — | · |
| 7 | Seedance 2.5 | 1474 | — | · |
| 8 | MiniMax H3 | 1460 | — | · |
| 9 | Muse Video | 1456 | — | · |
| 10 | HappyHorse 1.0 | 1427 | — | · |
| 11 | Sora 2 Pro | 1368 | — | · |
| 12 | Veo 3.1 | 1364 | — | · |
Image boards
Image Edit: LMArena
Tracking since Oct 19, 2026 · 2 snapshots · board dated 2026-10-06 → 2026-10-19 · lmarena.ai ↗
- Muse Image moved from #8 to #10 (2026-10-06 → 2026-10-19)
- ChatGPT Image Latest moved from #13 to #15 (2026-10-06 → 2026-10-19)
- Gemini Nano Banana 2.1 moved from #6 to #5 (2026-10-06 → 2026-10-19)
- MAI-Image-2.6 moved from #5 to #6 (2026-10-06 → 2026-10-19)
- Seedream 5.0 Pro moved from #10 to #11 (2026-10-06 → 2026-10-19)
- Grok Imagine Image (Quality) moved from #11 to #12 (2026-10-06 → 2026-10-19)
- Krea 3 Ultra joined at #8
- Ideogram 4.5 left the board (was #20)
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | GPT-Image-2.5 Sunburst | 1514 | = | |
| 2 | GPT-Image-2.5 Flare | 1481 | = | |
| 3 | GPT-Image-2 (Medium) | 1461 | = | |
| 4 | Grok Imagine Image 2.0 | 1439 | = | |
| 5 | Gemini Nano Banana 2.1 | 1426 | ▲ 1 | |
| 6 | MAI-Image-2.6 | 1425 | ▼ 1 | |
| 7 | Grok Imagine Image 2.0 (20260801) | 1415 | = | |
| 8 | Krea 3 Ultra | 1411 | new | · |
| 9 | MAI-Image-2.5 | 1400 | = | |
| 10 | Muse Image | 1398 | ▼ 2 | |
| 11 | Seedream 5.0 Pro | 1391 | ▼ 1 | |
| 12 | Grok Imagine Image (Quality) | 1391 | ▼ 1 | |
| 13 | Gemini 3 Pro Image (2K) | 1390 | ▼ 1 | |
| 14 | Gemini 3.1 Flash Image | 1387 | = | |
| 15 | ChatGPT Image Latest | 1387 | ▼ 2 |
Top 15 of 20 shown. Every snapshot keeps the full board.
1 model no longer on this board (kept for history)
- Ideogram 4.5: last seen 2026-10-06 at #20
Text-to-Image: Artificial Analysis
Tracking since Oct 19, 2026 · 2 snapshots · board dated 2026-10-17 → 2026-10-23 · artificialanalysis.ai ↗
- MAI-Image-2.6 moved from #6 to #5 (2026-10-17 → 2026-10-23)
- Grok Imagine Image 2.0 moved from #5 to #6 (2026-10-17 → 2026-10-23)
- MAI-Image-2.5 moved from #11 to #12 (2026-10-17 → 2026-10-23)
- FLUX 3 Image moved from #12 to #13 (2026-10-17 → 2026-10-23)
- Nano Banana 2 (Gemini 3.1 Flash Image) joined at #7
- Nano Banana Pro (Gemini 3 Pro Image) joined at #11
- Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) joined at #15
- Nano Banana 2 left the board (was #7)
- Nano Banana Pro left the board (was #13)
- Nano Banana 2 Lite left the board (was #15)
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | GPT Image 2.5 Sunburst (max) | 1222 | = | |
| 2 | GPT Image 2.5 Flare (max) | 1207 | = | |
| 3 | GPT Image 2 (high) | 1186 | = | |
| 4 | Nano Banana 2.1 | 1176 | = | |
| 5 | MAI-Image-2.6 | 1166 | ▲ 1 | |
| 6 | Grok Imagine Image 2.0 | 1166 | ▼ 1 | |
| 7 | Nano Banana 2 (Gemini 3.1 Flash Image) | 1140 | new | · |
| 8 | Muse Image | 1127 | = | |
| 9 | GPT Image 1.5 (high) | 1123 | = | |
| 10 | MAI-Image-2.6-Flash | 1119 | = | |
| 11 | Nano Banana Pro (Gemini 3 Pro Image) | 1118 | new | · |
| 12 | MAI-Image-2.5 | 1118 | ▼ 1 | |
| 13 | FLUX 3 Image | 1117 | ▼ 1 | |
| 14 | MAI-Image-2.5-Pro | 1114 | = | |
| 15 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 1107 | new | · |
Top 15 of 20 shown. Every snapshot keeps the full board.
3 models no longer on this board (kept for history)
- Nano Banana 2: last seen 2026-10-17 at #7
- Nano Banana Pro: last seen 2026-10-17 at #13
- Nano Banana 2 Lite: last seen 2026-10-17 at #15
Text-to-Image: LMArena
Tracking since Oct 19, 2026 · 1 snapshot · board dated 2026-10-07 · lmarena.ai ↗
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | GPT-Image-2.5 Sunburst | 1425 | — | · |
| 2 | GPT-Image-2.5 Flare | 1398 | — | · |
| 3 | GPT-Image-2 (Medium) | 1383 | — | · |
| 4 | Grok Imagine Image 2.0 | 1334 | — | · |
| 5 | MAI-Image-2.6 | 1332 | — | · |
| 6 | Gemini Nano Banana 2.1 | 1328 | — | · |
| 7 | Reve 2.1 | 1302 | — | · |
| 8 | Grok Imagine Image 2.0 (20260801) | 1294 | — | · |
| 9 | Muse Image | 1274 | — | · |
| 10 | Reve 2.0 | 1269 | — | · |
| 11 | Gemini 3.1 Flash Image | 1262 | — | · |
| 12 | Seedream 5.0 Pro | 1255 | — | · |
| 13 | Qwen-Image-3.0-Pro | 1255 | — | · |
| 14 | MAI-Image-2.5 | 1254 | — | · |
| 15 | Gemini 3.1 Flash-Lite Image | 1250 | — | · |
Top 15 of 20 shown. Every snapshot keeps the full board.
Audio boards
Music with vocals: AA-Music-Vocal v1.1
Tracking since Oct 19, 2026 · 1 snapshot · board dated 2026-10-17 · artificialanalysis.ai ↗
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | Suno v6 | 1146 | — | · |
| 2 | Suno v6-mini | 1130 | — | · |
| 3 | Mureka V9.5 | 1107 | — | · |
| 4 | Mureka V9 | 1057 | — | · |
| 5 | Lyria 3.5 | 1052 | — | · |
| 6 | Lyria 3 Pro | 1051 | — | · |
| 7 | Eleven Music v2.5 | 1012 | — | · |
| 8 | MiniMax Music 3.0 | 1000 | — | · |
| 9 | Eleven Music v2 | 999 | — | · |
| 10 | Eleven Music | 999 | — | · |
| 11 | MiniMax Music 2.6 | 998 | — | · |
| 12 | MiniMax Music 2.5+ | 985 | — | · |
| 13 | Melodia v3 | 972 | — | · |
Open-weights TTS (AA subset)
Tracking since Oct 19, 2026 · 2 snapshots · board dated 2026-10-17 → 2026-10-23 · artificialanalysis.ai ↗
- Step Audio EditX (Mar 2026) moved from #25 to #41 (2026-10-17 → 2026-10-23)
- Higgs Audio V3 TTS moved from #61 to #70 (2026-10-17 → 2026-10-23)
- Kokoro 82M v1.0 moved from #49 to #57 (2026-10-17 → 2026-10-23)
- Chatterbox moved from #72 to #78 (2026-10-17 → 2026-10-23)
- Fish Audio Drama 3 Preview moved from #6 to #9 (2026-10-17 → 2026-10-23)
- Maya1 moved from #65 to #68 (2026-10-17 → 2026-10-23)
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 9 | Fish Audio Drama 3 Preview | 1248 | ▼ 3 | |
| 11 | Breeze TTS 2 | 1225 | ▲ 2 | |
| 37 | Fish Audio S2 Pro | 1112 | ▲ 1 | |
| 41 | Step Audio EditX (Mar 2026) | 1099 | ▼ 16 | |
| 45 | Voxtral TTS | 1091 | ▲ 2 | |
| 57 | Kokoro 82M v1.0 | 1065 | ▼ 8 | |
| 58 | Magpie-Multilingual 357M (Feb 2026) | 1063 | ▼ 1 | |
| 68 | Maya1 | 1047 | ▼ 3 | |
| 69 | OpenAudio S1 Mini | 1044 | ▼ 3 | |
| 70 | Higgs Audio V3 TTS | 1044 | ▼ 9 | |
| 78 | Chatterbox | 1024 | ▼ 6 | |
| 81 | Zonos-v0.1 | 1000 | ▼ 3 |
Text-to-Speech: Artificial Analysis
Tracking since Oct 19, 2026 · 2 snapshots · board dated 2026-10-17 → 2026-10-23 · artificialanalysis.ai ↗
- StepAudio 2.5 TTS (Aug 2026) moved from #24 to #14 (2026-10-17 → 2026-10-23)
- Eleven v3 moved from #14 to #21 (2026-10-17 → 2026-10-23)
- v3 Conversational moved from #10 to #15 (2026-10-17 → 2026-10-23)
- Gemini 3.8 Flash-Lite TTS moved from #4 to #8 (2026-10-17 → 2026-10-23)
- Sonic 3.6 moved from #1 to #4 (2026-10-17 → 2026-10-23)
- Realtime TTS-2 moved from #9 to #6 (2026-10-17 → 2026-10-23)
- Voxtral 3 TTS joined at #12
- Ming-3.0-tts joined at #19
- Speech 2.8 HD joined at #24
- Azure HD 2.5 left the board (was #19)
- Simba 3.0 left the board (was #23)
- Step Audio EditX (Mar 2026) left the board (was #25)
| # | Model | Elo / arena score | Change | History |
|---|---|---|---|---|
| 1 | Eleven v4 Turbo | 1335 | ▲ 1 | |
| 2 | Eleven v4 | 1329 | ▲ 1 | |
| 3 | Qwen-Audio-3.1-TTS-Plus | 1301 | ▲ 2 | |
| 4 | Sonic 3.6 | 1289 | ▼ 3 | |
| 5 | Gemini 3.8 Flash TTS | 1278 | ▲ 2 | |
| 6 | Realtime TTS-2 | 1267 | ▲ 3 | |
| 7 | Qwen-Audio-3.0-TTS-Plus | 1265 | ▲ 1 | |
| 8 | Gemini 3.8 Flash-Lite TTS | 1251 | ▼ 4 | |
| 9 | Fish Audio Drama 3 Preview | 1248 | ▼ 3 | |
| 10 | Simba 3.2 | 1245 | ▲ 2 | |
| 11 | Breeze TTS 2 | 1225 | ▲ 2 | |
| 12 | Voxtral 3 TTS | 1218 | new | · |
| 13 | Realtime TTS-2 Flash | 1218 | ▼ 2 | |
| 14 | StepAudio 2.5 TTS (Aug 2026) | 1216 | ▲ 10 | |
| 15 | v3 Conversational | 1214 | ▼ 5 |
Top 15 of 25 shown. Every snapshot keeps the full board.
3 models no longer on this board (kept for history)
- Azure HD 2.5: last seen 2026-10-17 at #19
- Simba 3.0: last seen 2026-10-17 at #23
- Step Audio EditX (Mar 2026): last seen 2026-10-17 at #25