Leaderboard / Model
GPT-6 Sol
by OpenAIAssistantChatGPT Work and Codex on paid plans; API $2 / $10 per million input / output tokensUpdated 27 Sept 2026
GPT-6 Sol is OpenAI's GPT-6 model for coding and agents, released September 22, 2026 at half GPT-5.6 Sol's price. In Blender it writes Blender Python or drives your scene over MCP. It entered Season 3 at xhigh reasoning effort, so its 3D results are judged blind against GPT-6 Astra, Claude Opus 5.5, Grok 4.7 and the rest.
How to use GPT-6 Sol with Blender
Ask it for a bpy script and run it in Blender's Scripting workspace, or connect Codex (or another MCP-capable client) running gpt-6-sol to MCP for Blender for live scene access.
Best for: Scripted and MCP-driven Blender work where you want GPT-6-class coding at half GPT-5.6 Sol's price; the blind votes will show how its 3D output compares.
Which AI built it better?
Two random Season 3 entrants answered the same prompt. Pick the better build and the names are revealed. Your vote counts toward the leaderboard.
Picking a matchup…
What GPT-6 Sol built
All 12 Season 3 prompts, one script each, exactly as it came out. Click a build to orbit it in 3D, or compare every AI on the same prompt.
Treasure chestProps Desk lampProps Low-poly knightCharacters Mushroom creatureCharacters Stone watchtowerArchitecture Modern houseArchitecture Pickup truckVehicles Sailing boatVehicles Potted plantNature Island sceneNature Spiral staircaseAbstract Chess setAbstract
Earlier seasons stay published in the Season 1 archive.
Arena record
Season 3Calibrating
- 1
- Entrants
- 12
- Builds
- of 12 prompts
- 29
- Times voted on
- GPT-6 SolCalibrating1591 ±7918–10–1
Every entrant starts each season at 1500 ±350 and stays unranked until enough votes are in. Nothing carries over between seasons, ratings move only on real votes, and every ballot is published. How the rating works
What is GPT-6 Sol?
GPT-6 Sol is OpenAI's GPT-6 model built for complex coding and agentic workflows, released on September 22, 2026 alongside GPT-6 Luna and trained with methods similar to GPT-6 Astra. It has a 1.05M-token context window, up to 128K output tokens, an April 20, 2026 knowledge cutoff, and reasoning effort from none up to max (default medium).
On the API (model ID gpt-6-sol) it costs $2 per million input tokens, $0.20 cached and $10 output, half of GPT-5.6 Sol's $4 / $20. OpenAI says it makes about half as many factual mistakes as GPT-5.6 Sol. It is in ChatGPT Work and Codex on Plus, Pro, Business, Enterprise and Edu.
It has no first-party Blender integration: it writes Blender Python (bpy) you run in the Scripting workspace, or drives a live scene through an MCP-capable client such as Codex connected to MCP for Blender. It has its own profile, apart from ChatGPT and GPT-6 Astra, because the arena competes per model: GPT-5.6 Sol keeps its own record.
Sol joined Season 3 on September 27, 2026, at xhigh reasoning effort, writing a single bpy script per prompt executed once in headless Blender 5.1. All twelve prompts produced geometry, with no script errors.
GPT-6 Sol at a glance
- Released: September 22, 2026
- API model ID: gpt-6-sol
- Price: $2 per million input tokens, $0.20 cached input, $10 output (GPT-5.6 Sol: $4 / $20)
- Context window: 1.05M tokens; max output: 128K tokens
- Reasoning effort: none, low, medium (default), high, xhigh, max
- Knowledge cutoff: April 20, 2026
- Available in ChatGPT Work and Codex on Plus, Pro, Business, Enterprise and Edu, and in the API
GPT-6 Sol vs GPT-5.6 Sol and GPT-6 Astra
OpenAI prices Sol at half of GPT-5.6 Sol and says it makes about half as many factual mistakes. Its launch figures put Sol at 33.2% on AutomationBench at xhigh, ahead of GPT-6 Astra at low effort at a fraction of the cost, 60.5% on OSWorld 2.0 offline at xhigh, and 68.8% on DeepSWE v1.1 at max. OpenAI still calls Astra its best model across the board.
None of those measure 3D. GPT-5.6 Sol has a full arena record and still competes in Season 3 on the files it was judged on before; GPT-6 Sol ran the same twelve prompts under the same rule, one script per prompt executed once in headless Blender 5.1, at xhigh effort like the other Season 3 newcomers. The blind votes decide whether the generation step shows up in the geometry.
Arena status
GPT-6 Sol joined Season 3 on September 27, 2026, at xhigh reasoning effort, the same setting as Claude Opus 5.5 and Grok 4.7. All twelve prompts produced geometry with no script errors, and every output is on this page exactly as generated. It entered blind voting at the Glicko-2 default rating, like every other competitor, and stays unranked until it clears the calibration floors.
Strengths and limitations
Strengths
- Strength:OpenAI's launch results: 33.2% on AutomationBench at xhigh (ahead of GPT-6 Astra at low effort, at a fraction of the cost) and 68.8% on DeepSWE v1.1 at max
- Strength:Half the price of GPT-5.6 Sol: $2 / $10 per million tokens
- Strength:1.05M-token context and 128K output tokens, room for long bpy scripts and scene sessions
- Strength:Exported geometry on all twelve arena prompts with no script errors, with long procedural scripts building up to 225 objects per scene
Limitations
- Limitation:No blind-vote record yet: it joined Season 3 on September 27, 2026, so its rating is uncalibrated and nothing here is a quality claim
- Limitation:OpenAI still calls GPT-6 Astra its best model across the board; Sol is the cheaper tier
- Limitation:Not a mesh generator: geometry comes from code, which makes organic and sculpted subjects the structural weak spot
- Limitation:Higher effort settings cost more time and tokens per generation than the default
GPT-6 Sol and Blender: FAQ
When was GPT-6 Sol released?
OpenAI released GPT-6 Sol on September 22, 2026, together with GPT-6 Luna. It is available in ChatGPT Work and Codex on paid plans and in the API as gpt-6-sol.
How much does GPT-6 Sol cost?
$2 per million input tokens, $0.20 for cached input and $10 per million output tokens: half of GPT-5.6 Sol's $4 / $20. OpenAI says these are permanent prices.
Can GPT-6 Sol use Blender?
Yes, through the same two routes as other assistants: it writes Blender Python you run in the Scripting workspace, or Codex (or another MCP-capable client) running it connects to MCP for Blender and works in a live scene. There is no official OpenAI Blender connector.
Is GPT-6 Sol better than GPT-5.6 Sol at 3D modelling?
Not settled yet. OpenAI's launch results favour it on coding and agent benchmarks, but none of them measure 3D. Both are in Season 3 on the same twelve prompts, and the blind votes will answer it.