Sonnet 5.5 vs Sol 6 on same prompts same price point

Did Anthropic found some black magic or why the hell are their new models so amazing?! Are they completely based on model 2 or what is going on? Holy moly.

Both are slop 🤣 like video models from 2024; probably by the end of next year, computer science in general will be solved.

Anthropic finally making a comeback

a this rate sol will be competing against haiku on low

Devday or else it will be a long while of clude dominance

Now this is INSANE ! For a sonnet model 🤯

Watched the comparison. Sol 6 still wins on long-horizon planning; Sonnet feels snappier on the short loops. Same price point makes the trade-off clearer than the marketing.

Can you check how Sonnet handles creative writing?

this shit is gonna be stronger than astra and fable is gonna change the world, openai only has a chance with bel to compete

Sonnet looks crazy 🤯

Feel so sorry for anyone trying to build ui with oai models.

Sonnett 5.5 is far ahead. It's all Anthropic, lets see what OpenAI Devday brings on Tuesday. The consumer is loving it.

Give me the prompt please?

finally, a benchmark with a budget column

how much estimated money was spent by one and the other. obviously sonnet spent a lot more since it worked for 90 minutes +?

I’m confused is sonnet 5.5 out?

Sonnet is like way ahead 🔥

My Sol ain't even doing all that

How does the latency hold up once you push past a few tool calls in a row? Sol 6 usually drifts when context stacks up, curious if Sonnet 5.5 stays snappy there.

Could you share the prompts too? Would be useful to rerun this as both models get updated.