We ran Muse Spark 1.3 on our Cybersecurity benchmark, and it's actually not that good (yet) 😬
At pass@1, it rediscovers an average of 19/32 CVEs. In comparison, Grok 4.6 scores 23.3/32 - When pooling the results of 3 runs (pass@3), Muse scores 24/32. DeepSeek V4 Pro gets
I tried it with the same prompt, but to be honest, DeepSeek-V4-Flash had a better pop-up level.
It's too much to exceed GPT-5.6 Sol Muse Spark1.3💦
How to steal Claude is a question, not a hack.
On September 3, CNBC interviewed Jacob Klein, who catches unauthorized access at Anthropic, and said the method is as follows. Tens to hundreds of thousands of fake accounts are created using stolen credit cards and credentials, including stolen Anthropic accounts. Ask Claude a million questions on that account…
DGX Spark x 2 + DeepSeek V4 Flash Vision Exp I can now reach this speed for the time being.
KV cache has been switched to FP8. The maximum context is limited to 512K. It's about two steps away from the speed of 0731.
You have reached the end of the archive
All of deepseek