DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
1–10 of 485 posts
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#2Pretty amazing that a relatively small Chinese hedge fund can build AI better than almost anyone.
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#33.2-Exp came out in September: this is 3.2, along with a special checkpoint (DeepSeek-V3.2-Speciale) for deep reasoning that they're claiming surpasses GPT-5 and matches Gemini 3.0
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#4Benchmarks are super impressive, as usual. Interesting to note in table 3 of the paper (p. 15), DS-Speciale is 1st or 2nd in accuracy in all tests, but has much higher token output (50% more, or 3.5x vs gemini 3 in the codeforces test!).
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#5I genuinely do not understand the evaluations of the US AI industry. The chinese models are so close and far cheaper
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#6Benchmarks are super impressive, as usual. Interesting to note in table 3 of the paper (p. 15), DS-Speciale is 1st or 2nd in accuracy in all tests, but has much higher token output (50% more, or 3.5x vs gemini 3 in the codeforces test!).
The higher token output is not by accident. Certain kinds of logical reasoning problems are solved by longer thinking output. Thinking chain output is usually kept to a reasonable length to limit latency and cost, but if pure benchmark performance is the goal you can crank that up to the max until the point of diminishing returns. DeepSeek being 30x cheaper than Gemini means there’s little downside to max out the thinking time. It’s been shown that you can further scale this by running many solution attempts in parallel with max thinking then using a model to choose a final answer, so increasing reasoning performance by increasing inference compute has a pretty high ceiling.
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#7I genuinely do not understand the evaluations of the US AI industry. The chinese models are so close and far cheaper
Yet tbh if the US industry had not moved ahead and created the race with FOMO it would not had been easier for Chinese strategy to work either.
The nature of the race may change as yet though, and I am unsure if the devil is in the details, as in very specific edge cases that will work only with frontier models ?
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#8I genuinely do not understand the evaluations of the US AI industry. The chinese models are so close and far cheaper
Valuation is not based on what they have done but what they might do, I agree tho it's investment made with very little insight into Chinese research. I guess it's counting on deepseek being banned and all computers in America refusing to run open software by the year 2030 /snark
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#9It's awesome that stuff like this is open source, but even if you have a basement rig with 4 NVIDIA GeForce RTX 5090 graphic cards ($15-20k machine), can it even run with any reasonable context window that isn't like a crawling 10/tps?
Frontier models are far exceeding even the most hardcore consumer hobbyist requirements. This is even further
Re: DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]
#10I genuinely do not understand the evaluations of the US AI industry. The chinese models are so close and far cheaper
It's all about the hardware and infrastructure. If you check OpenRouter, no provider offers a SOTA chinese model matching the speed of Claude, GPT or Gemini. The chinese models may benchmark close on paper, but real-world deployment is different. So you either buy your own hardware in order to run a chinese model at 150-200tps or give up an use one of the Big 3.
The US labs aren't just selling models, they're selling globally distributed, low-latency infrastructure at massive scale. That's what justifies the valuation gap.
Edit: It looks like Cerebras is offering a very fast GLM 4.6