Institute Of Foundation Models vs Anthropic vs Microsoft vs Google

Head to head on 26 boards: Anthropic places best on 15, Google places best on 7, Microsoft places best on 4 and Institute Of Foundation Models places best on 0. All 4 ranked on 0 of 28 boards.

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

IOInstitute Of Foundation ModelsAnthropicMicrosoftGoogle

Board	Read	IOInstitute Of Foundation Models	Anthropic	Microsoft	Google
Chat2
Text ArenaArena ratingtodaytoday	Oct 6todaytoday		#21,505Claude Opus 4.6		#11,525best of the pickedGemini 4 Argon
MultiChallengeScoretodaytoday	Oct 6todaytoday		#1259.0%Claude Opus 4.5		#371.4%best of the pickedGemini 3.1 Pro Preview
Reasoning7
AA IntelligenceIndextodaytoday	Oct 6todaytoday	#2830.5K2 Horizon 375B A23B	#157.6best of the pickedClaude Opus 5.5		#552.6Gemini 4 Argon
MMLU-ProAccuracy · staletodaytoday	Oct 6todaytoday		#489.1%Claude Opus 4.6	#7576.0%Phi 4 Plus	#191.2%best of the pickedGemini 3.1 Pro Preview
Epoch ECIECItodaytoday	Oct 6todaytoday		#1167.3best of the pickedClaude Opus 5.5		#16157.3Gemini 3.7 Flash
LiveBenchGlobal averagetodaytoday	Oct 6todaytoday		#183.4%best of the pickedClaude Fable 5.1		#1478.8%Gemini 3.7 Flash
HLEAccuracytodaytoday	Oct 6todaytoday		#246.5%best of the pickedClaude Fable 5.1		#346.4%Gemini 3.1 Pro Preview
GPQA DiamondAccuracytodaytoday	Oct 6todaytoday		#295.6%best of the pickedClaude Sonnet 5.5		#395.4%Gemini 3.8 Flash
ARC-AGI-2Scoretodaytoday	Oct 6todaytoday		#393.3%best of the pickedClaude Opus 5.5		#889.2%Gemini 3.8 Flash
Coding4
WebDev ArenaArena ratingtodaytoday	Oct 6todaytoday		#11,815best of the pickedClaude Opus 5.5		#81,680Gemini 4 Argon
SWE-Bench ProResolvedtodaytoday	Oct 6todaytoday		#198.0%best of the pickedClaude Opus 5		#958.8%Gemini 3.8 Flash
SWE-bench VerifiedResolved · staletodaytoday	Oct 6todaytoday		#176.8%best of the pickedClaude Opus 4.5		#275.8%Gemini 3 Flash
Aider PolyglotPass rate · staletodaytoday	Oct 6todaytoday		#1072.0%Claude Opus 4		#383.1%best of the pickedGemini 2.5 Pro (Jun 2025)
Agents and tools3
BFCLOverall accuracytodaytoday	Oct 6todaytoday		#177.5%best of the pickedClaude Opus 4.5	#5028.8%Phi 4	#372.5%Gemini 3 Pro
MCP AtlasPass ratetodaytoday	Oct 6todaytoday		#287.2%best of the pickedClaude Fable 5.1		#683.6%Gemini 3.5 Flash
Search ArenaArena ratingtodaytoday	Oct 6todaytoday		#21,253best of the pickedClaude Opus 4.6		#91,210Gemini 3.1 Pro Grounding
Vision2
Vision ArenaArena ratingtodaytoday	Oct 6todaytoday		#11,309best of the pickedClaude Fable 5		#51,296Gemini 3.7 Flash
Document ArenaArena ratingtodaytoday	Oct 6todaytoday		#11,516best of the pickedClaude Opus 5		#161,463Gemini 3.5 Flash
Image2
Text-to-ImageArena ratingtodaytoday	Oct 6todaytoday			#51,333best of the pickedMAI Image 2.6	#91,261Gemini 3.1 Flash Image
Image EditArena ratingtodaytoday	Oct 6todaytoday			#51,427best of the pickedMAI Image 2.6	#101,390Gemini 3 Pro Image
Video2
Text-to-VideoArena ratingtodaytoday	Oct 6todaytoday				#11,516Gemini Omni 1.1 Flash
Image-to-VideoArena ratingtodaytoday	Oct 6todaytoday				#21,488Gemini Omni 1.1 Flash
Speech2
Open ASRAverage WERtodaytoday	Oct 6todaytoday			#23.81 WERbest of the pickedazure-speech-07-2026 (microsoft)	#629.51 WERgemma-4-E4B-it (google)
Speech ArenaArena ratingtodaytoday	Oct 6todaytoday			#281,131Azure Hd 2.5	#51,275best of the pickedGemini 3.8 Flash TTS
Embeddings1
MTEBMean task scoretodaytoday	Oct 6todaytoday			#174.3%best of the pickedharrier-oss-v1-27b (microsoft)	#568.4%gemini-embedding-001 (google)
Speed1
AA SpeedOutput speedtodaytoday	Oct 6todaytoday	#7124 tok/sK2 Horizon 375B A23B	#5139 tok/sClaude Sonnet 5.5		#1249 tok/sbest of the pickedGemini 3.8 Flash
Price1
AA PriceBlended pricetodaytoday	Oct 6todaytoday		#17$4.00Claude Sonnet 5.5		#10$1.50best of the pickedGemini 3.8 Flash
Usage1
OpenRouterWeekly tokenstodaytoday	Oct 6todaytoday		#122.6T tokbest of the pickedClaude Opus 5.5		#132.2T tokGemini 3.8 Flash

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.