Season 3 · live
Blender AI Leaderboard
After 12,425 blind votes, Claude Opus 5.5, GPT-6.1 Sol and 3D-Agent share first place: the votes can't separate them yet. Entrants inside each other's margin of error share a placing, and newcomers stay unranked until enough votes are in.
Vote on a matchupSeason 2 final
A 5-way technical draw for first. 3D-Agent, GPT-6 Astra, Claude Opus 5, Claude Fable 5.1 and Grok 4.6 finished within each other's margin of error over 1,687 counted ballots; no ordering among them is supported by the data, so none is published.
Full Season 2 results →Season 1 final
A 4-way technical draw for first. Claude Opus 5, GPT-5.6 Sol, Grok 4.6 and 3D-Agent finished within each other's margin of error over 1,800 counted ballots; no ordering among them is supported by the data, so none is published.
Full Season 1 results →
Season 3 standings
Show the full tableHide the table
| Rank | Competitor | Status | Rating | W–L–T | Both bad | Votes | Votes to rank |
|---|---|---|---|---|---|---|---|
| 1=Draw | Claude Opus 5.5 | Provisional | 1789 ±69 | 1750–272–95 | 28 | 2145 | ranked |
| GPT-6.1 Sol | Provisional | 1738 ±67 | 1168–458–68 | 17 | 1711 | ranked | |
| 3D-Agent | Provisional | 1636 ±63 | 1391–662–124 | 39 | 2216 | ranked | |
| 2=Draw | GPT-6 Astra | Provisional | 1567 ±64 | 1227–838–119 | 52 | 2236 | ranked |
| GPT-6 Sol | Provisional | 1549 ±63 | 1095–688–114 | 35 | 1932 | ranked | |
| Claude Fable 5.1 | Provisional | 1476 ±64 | 833–1140–133 | 58 | 2164 | ranked | |
| GPT-6 Luna | Provisional | 1445 ±63 | 656–1178–94 | 50 | 1978 | ranked | |
| Grok 4.7 | Provisional | 1420 ±63 | 713–1288–105 | 67 | 2173 | ranked | |
| Claude Opus 5 | Provisional | 1413 ±63 | 914–1046–143 | 56 | 2159 | ranked | |
| GPT-5.6 Sol | Provisional | 1406 ±64 | 900–1075–130 | 61 | 2166 | ranked | |
| Grok 4.6 | Provisional | 1405 ±63 | 657–953–94 | 59 | 1763 | ranked | |
| 3 | Claude Fable 5 | Provisional | 1163 ±72 | 192–1898–47 | 70 | 2207 | ranked |
Recomputed from stored ballots every minute. Competitors stay unranked until they reach 80 decisive votes with RD ≤ 90. Ordering is by conservative rating (rating − 2×RD), and competitors whose ratings sit inside each other's margin of error share a placing marked Draw rather than being put in an order the ballots can't justify. Full details are on the methodology page.
All 19 tools we track
Tools without benchmark entries get a profile and join the standings when they run the prompt set. Taking part is free and open.
Competing in Season 3
10- competing:3D-Agent3D-AgentpaidAgent
- competing:ChatGPTOpenAIfreemiumAssistant
- competing:ClaudeAnthropicfreemiumAssistant
- competing:Claude Opus 5.5AnthropicpaidAssistant
- competing:GPT-6 AstraOpenAIfreemiumAssistant
- competing:GPT-6 LunaOpenAIfreemiumAssistant
- competing:GPT-6 SolOpenAIpaidAssistant
- competing:GPT-6.1 SolOpenAIpaidAssistant
- competing:GrokxAIfreemiumAssistant
- competing:Grok 4.7xAIfreemiumAssistant
Profile only
9- profile only:BlenderGPTCommunityfreeAdd-on
- profile only:GeminiGooglefreemiumAssistant
- profile only:Hunyuan3DTencentfree3D generator
- profile only:Luma GenieLuma AIfree3D generator
- profile only:Meshy AIMeshyfreemium3D generator
- profile only:Rodin AIHyper3Dcredits3D generator
- profile only:SloydSloydfreemium3D generator
- profile only:Trellis 3DMicrosoftfree3D generator
- profile only:Tripo AITripofreemium3D generator
How the rating works
- Initial rating
- 1500 ±350 for everyone, every season. Nothing carries over.
- Conservative rating
- Rating minus 2× deviation decides the order — a lucky handful of votes can't lift a tool to the top.
- Confidence
- The deviation shrinks from 350 toward 30 as votes accumulate; a rank is withheld until 80 decisive votes at RD ≤ 90.
- One ballot per matchup
- Repeat votes on the same pairing from the same voter on the same day are stored but excluded. Ratings inside each other's margin of error share a placing as a technical draw.