@teortaxesTex' DeepSeek V4 predictions thread
7
1.9kṀ2261
Jan 1
37%
>=1.5T parameters
59%
>=52B active parameters
61%
>=25T pretraining tokens
51%
uses some non-AdamW optimizer
35%
DS-MoE with adaptative expert count
41%
intra-expert communication
51%
>=512 experts
52%
>=16 active experts
59%
>= 2 shared experts
72%
Some variation of NSA (Native Sparse Attention)
49%
1M+ Context
22%
Gemini 2.5 Pro tier or higher on FictionBench (90.6%+ at 192k)
15%
>= 44% on Humanity's Last Exam (text only) at scale.com leaderboard
16%
>= 73% on SWE-Bench Verified (according to epoch.ai)
21%
>= 60% on BrowseComp (https://www.kaggle.com/benchmarks/openai/browsecomp)
29%
>= 50% on TerminalBench (https://www.tbench.ai/leaderboard)
39%
Some image input (multimodality)
20%
DeepSeek reports some results with a full-blown deep research agent, and emphasizes that this is the intended use-mode

Teortaxes gave some point estimates. These are not as amenable to prediction market forecasting so I turned them into over/under forecasts. I may add forecasts from other commenters in the thread later on, so these may not only be forecasts by Teo

See post for more (including forecasts I wasn't able to turn into market options):

Unranked
Get
Ṁ1,000
to start trading!

People are also trading

Sort by:

Would Deepseek 3.2 DSA (derpseek sparse attention) in v4 resolve as “some variant of NSA” for the purposes of this market?

@ookina_inu hmmm dunno the details enough to evaluate this. i'd default to asking teo maybe. if you know the details of both DSA and NSA and have an opinion one way or another lmk

@Bayesian Gotcha. I honestly think this could go either way. Seems sufficiently different from NSA to not literally be NSA, but plausibly could fit in “some variant of.” Will update if I form a stronger opinion

deepseek r1 does 5.7%. but current frontier (Droid (claude-opus-4-1)) gets 58.8%

People are also trading

© Manifold Markets, Inc.•Terms•Privacy