K2 Horizon 3.7B vs Claude Fable 5.1 vs Qwen3.8 Max (0902)

• IOK2 Horizon 3.7B
• Claude Fable 5.1
• Qwen3.8 Max (0902)

Add a model

Add:GPT-6 AstraGemini 3.8 FlashMuse SparkKimi K3

Head to head on 8 boards: Claude Fable 5.1 places best on 6, Qwen3.8 Max (0902) places best on 2 and K2 Horizon 3.7B places best on 0. All 3 ranked on 1 of 14 boards.

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

IOK2 Horizon 3.7BClaude Fable 5.1Qwen3.8 Max (0902)

Board	Read	IOK2 Horizon 3.7B	Claude Fable 5.1	Qwen3.8 Max (0902)
Chat1
Text ArenaArena ratingtodaytoday	Oct 6todaytoday		#61,501best of the picked	#201,482
Reasoning6
AA IntelligenceIndextodaytoday	Oct 6todaytoday	#6615.6	#353.4best of the picked	#1045.4
LiveBenchGlobal averagetodaytoday	Oct 6todaytoday		#183.4%best of the picked	#1578.5%
Epoch ECIECItodaytoday	Oct 6todaytoday		#5164.7best of the picked	#22156.4
HLEAccuracytodaytoday	Oct 6todaytoday		#246.5%
ARC-AGI-2Scoretodaytoday	Oct 6todaytoday		#690.0%
GPQA DiamondAccuracytodaytoday	Oct 6todaytoday			#2192.7%
Coding2
WebDev ArenaArena ratingtodaytoday	Oct 6todaytoday		#51,749best of the picked	#91,671
SWE-Bench ProResolvedtodaytoday	Oct 6todaytoday		#292.2%
Agents and tools1
MCP AtlasPass ratetodaytoday	Oct 6todaytoday		#287.2%
Vision2
Vision ArenaArena ratingtodaytoday	Oct 6todaytoday		#131,288	#21,301best of the picked
Document ArenaArena ratingtodaytoday	Oct 6todaytoday		#21,513
Speed1
AA SpeedOutput speedtodaytoday	Oct 6todaytoday		#1568 tok/sbest of the picked	#2439 tok/s
Price1
AA PriceBlended pricetodaytoday	Oct 6todaytoday		#24$20.0	#14$3.00best of the picked

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.