gpt-transcribe (openai) vs Qwen3.8 Max (0902) vs Kimi K3

• gpt-transcribe (openai)
• Qwen3.8 Max (0902)
• Kimi K3

Add a model

Add:Claude Fable 5.1Gemini 3.8 FlashMuse SparkGLM-5.3

Head to head on 8 boards: Qwen3.8 Max (0902) places best on 4, Kimi K3 places best on 4 and gpt-transcribe (openai) places best on 0. All 3 ranked on 0 of 14 boards.

Two at a time on this screen. Tap one to swap it in.

gpt-transcribe (openai)Qwen3.8 Max (0902)Kimi K3

Board	Read	gpt-transcribe (openai)	Qwen3.8 Max (0902)	Kimi K3
Chat1
Text ArenaArena rating1dyesterday	Oct 61dyesterday		#201,482	#131,488best of the picked
Reasoning5
AA IntelligenceIndex1dyesterday	Oct 61dyesterday		#1045.4best of the picked	#1343.6
LiveBenchGlobal average1dyesterday	Oct 61dyesterday		#1578.5%	#1379.2%best of the picked
Epoch ECIECI1dyesterday	Oct 61dyesterday		#22156.4	#15157.4best of the picked
GPQA DiamondAccuracy1dyesterday	Oct 61dyesterday		#2192.7%	#1893.1%best of the picked
ARC-AGI-2Score1dyesterday	Oct 61dyesterday			#2960.4%
Coding2
WebDev ArenaArena rating1dyesterday	Oct 61dyesterday		#91,671best of the picked	#101,658
SWE-Bench ProResolved1dyesterday	Oct 61dyesterday			#488.2%
Agents and tools1
MCP AtlasPass rate1dyesterday	Oct 61dyesterday			#882.3%
Vision1
Vision ArenaArena rating1dyesterday	Oct 61dyesterday		#21,301
Speech1
Open ASRAverage WER1dyesterday	Oct 61dyesterday	#204.57 WER
Speed1
AA SpeedOutput speed1dyesterday	Oct 61dyesterday		#2439 tok/sbest of the picked	#2536 tok/s
Price1
AA PriceBlended price1dyesterday	Oct 61dyesterday		#14$3.00best of the picked	#22$6.00
Usage1
OpenRouterWeekly tokens1dyesterday	Oct 61dyesterday			#151.7T tok

Green marks the best placing among the picked where two or more are ranked; ties are marked for each. Ranks are each board's own order (lower is better), scores as the board published them. Dash: not on that board.