dovinmu

Analysis of Parameter Count Estimates for Frontier AI Models: Grok 3, Claude 3.7 Sonnet, and GPT-4.5

Recent advancements in large language models (LLMs) have sparked intense interest in understanding their architectural scale, particularly regarding parameter counts. This analysis synthesizes available evidence to estimate the parameter sizes of three frontier models: xAI's Grok 3, Anthropic's Claude 3.7 Sonnet, and OpenAI's GPT-4.5. While manufacturers typically withhold exact architectural details, multiple independent analyses and technical comparisons provide credible estimates.

Grok 3: Decoding xAI's "10x Power" Claim

Computational Power vs. Parameter Count

xAI's marketing claims about Grok 3 having "10x the computational power" of Grok 2[^4] require careful interpretation. The 100,000 Nvidia H100 GPUs mentioned in implementation details[^4] suggest substantial distributed training infrastructure rather than direct parameter scaling. Computational power metrics (FLOPs) correlate with but don

dovinmu

Get in touch at [email protected]

FTE: Lead Engineer

MiraDx (2021 - 2023)

MiraDx is a molecular diagnostics lab with specialty cancer tests. I was hired as the first in-house engineer, built and managed an in-house engineering team of 3, and worked closely with the founder/CEO to identify and decide direction of engineering needs.

I owned the Typescript+Nextjs platform and worked with UCLA scientists to productionize the statistics model in Python. I brought in security researchers to harden the codebase's security, redesigned the system for new markets, and productionized research classification models.

Project: health data pipeline

dovinmu

# https://gist.github.com/dovinmu/bcab247b063cd3e294e57d06440268a1

def popParens(s):
    return s[s.find('(')+1:s.rfind(')')]

def getArgs(s):
    def appendArg(arg):
        if '(' in arg:
            args.append(getArgs(arg.strip()))
        else:

dovinmu

def loadWordSet():
    s = set()
    try:
        with open('/usr/share/dict/american-english', 'r') as f:
            for line in f:
                s.add(line.strip())
    except:
        print("Could not find built-in dictionary")
        # rely on https://github.com/dwyl/english-words.git
        try: