Top Tweets for #chemcensor
2.6B parameters beating purpose-built models in single-step retrosynthesis! π
High time to ensemble these fast SOTA LLMs with search tree algorithms to power the ultimate multi-step retrosynthesis engine. #DDDBench #ChemCensor
πͺΆ What can a 2.6B-parameter language model achieve in retrosynthesis?
With a leak-proof, broad-coverage drug discovery benchmark #DDDBench, it becomes easier to evolve general-purpose models into SOTA-level specialists.
#insilicoSOTAFM

Weβve updated the #DDDBench with our own models, including LFM2-MMAI-Chem-SSRS-2.0, our single-step retrosynthesis specialist and the current SOTA on #ChemCensor.

BTW, our specialist model C3LM is SOTA on single step retrosynthesis! We share the treasure details from data for SFT and rewards for RFT
You only need 2.6B params for the task thanks to efficient @liquidai LFM architecture!
Also, our key simulation engine #ChemCensor is public
Iβm excited to share the follow-up paper to our #ChemCensor framework for evaluating single-step retrosynthesis π§ͺ
Our C3LM model was already competitive with, and in some cases superior to, general-purpose foundation models, but a gap remained between LLMs and conventional retrosynthesis models.
We finally closed and surpassed that gap π
The key ingredients wereβ¦ π
4/n
π Read the paper: https://t.co/JXVfb3PmCQ
π€ Join the discussion on Hugging Face: https://t.co/3YqsGCX1aP
π Recent #ChemCensor benchmark results on #DDDBench: https://t.co/5MmJq51CjX
Kudos to team @sumrexromanus @chem_logos @chemcensor @mathieu_reymond @VladAladin @AlexAliper @biogerontology
Iβm excited to share the follow-up paper to our #ChemCensor framework for evaluating single-step retrosynthesis π§ͺ
Our C3LM model was already competitive with, and in some cases superior to, general-purpose foundation models, but a gap remained between LLMs and conventional retrosynthesis models.
We finally closed and surpassed that gap π
The key ingredients wereβ¦ π
LFM2-2.6B was an excellent base model for our #ChemCensor paper! Looking forward to exploring what LFM2.5-2.6B can unlock with its improved capabilities.
Today we release LFM2.5-2.6B, an agentic model that runs entirely on-device. It plans, calls tools, and works through multi-step tasks on phones, laptops, PCs, and robots. Data never leaves the device, and the marginal cost of each run is essentially zero.
> Pre-trained on ~34T tokens
> LFM2.5 flagship hybrid architecture
> Context length: 128K
> Vocab size: 128K
> balanced intelligence per watt
> customizable on a single GPU for any specialized task
> LFM2 open-weight license
Comparable or better scores compared to models up to nearly 4x its size:
> ToolSandbox 77.83, ahead of Qwen3.5-9B at 76.44
> Multi-IF 80.07, ahead of Gemma-4-E4B-it at 77.35
> IFStruct 85.49, ahead of Qwen3.5-9B at 78.50
π§΅

At @InSilicoMeds, we built #URSA to benchmark retrosynthesis closer to how chemists evaluate routes. Kudos to @sumrexromanus !
#ChemCensor #ICML2026 evaluates reaction plausibility. URSA then scores full routes hierarchically using Solv-N metrics: from formal validity (Solv-0 π΄) to reaction-center logic (Solv-1 π ) and functional-group compatibility (Solv-2 π’). For more details: read our preprint: https://t.co/zY525e6B3n
Last Seen Hashtags on Sotwe
omegle ((((()))))******** filter:native_video
Seen from United States
mikalafuente
Seen from Argentina
tourismealapompe
Seen from United States
ει£γγγ£γΊγγΌ
Seen from United States
cumtrib
Seen from United States
ΩΩΩΨ³
Seen from Saudi Arabia
ΰΉΰΈΰΈ‘ΰΈͺΰΉΰΈΰΈΰΈΰΉΰΈ₯ΰΈΰΉ
Seen from United States
Torbe
Seen from Mexico
billford
Seen from Peru
ΰΈ£ΰΈΈΰΈ‘ΰΉΰΈ’ΰΉΰΈ
Seen from Thailand
Most Popular Users

Elon Musk 
@elonmusk
241.6M followers

Barack Obama 
@barackobama
119M followers

Cristiano Ronaldo 
@cristiano
114.2M followers

Donald J. Trump 
@realdonaldtrump
111.8M followers

Narendra Modi 
@narendramodi
107.2M followers

Rihanna 
@rihanna
98.7M followers

NASA 
@nasa
92.4M followers

Justin Bieber 
@justinbieber
91.8M followers

KATY PERRY 
@katyperry
89.9M followers

Taylor Swift 
@taylorswift13
83.8M followers

Lady Gaga 
@ladygaga
75.3M followers

Virat Kohli 
@imvkohli
73.2M followers

Kim Kardashian 
@kimkardashian
70.8M followers

YouTube 
@youtube
68.8M followers

Neymar Jr 
@neymarjr
66.2M followers

Bill Gates 
@billgates
65M followers

Selena Gomez 
@selenagomez
63M followers

The Ellen Show
@theellenshow
62.3M followers

CNN 
@cnn
61.8M followers

X 
@x
60.7M followers




