Thivakar Jeyaseelan

Thiv

1 86

AI & ML interests

None yet

Recent Activity

liked a dataset about 11 hours ago

HuggingFaceFW/fineweb

reacted to mlabonne's post with ❤️ about 13 hours ago

Deploy local agents everywhere with https://huggingface.co/LiquidAI/LFM2.5-2.6B Unlike agents that depend on cloud APIs, local agents give you free inference, low latency, and real privacy. Removing the per-token cost changes how developers build: agents can now be massively parallelized on local hardware, running background tasks that burn through millions of tokens at no marginal cost!

liked a model 1 day ago

mistralai/Shieldstral-1.0-3B

View all activity

Organizations

None yet

liked a dataset about 11 hours ago

HuggingFaceFW/fineweb

Viewer • Updated Jul 11, 2025 • 52.5B • 379k • 3.11k

reacted to mlabonne's post with ❤️ about 13 hours ago

Post

2866

Deploy local agents everywhere with LiquidAI/LFM2.5-2.6B

Unlike agents that depend on cloud APIs, local agents give you free inference, low latency, and real privacy.

Removing the per-token cost changes how developers build: agents can now be massively parallelized on local hardware, running background tasks that burn through millions of tokens at no marginal cost!

7 replies

liked 5 models 1 day ago

reacted to sergiopaniego's post with 🤗 2 days ago

Post

2509

LFM2.5-2.6B just dropped!

and the @liquidai blog comes with some nice details about the training procedure, so let's analyze it.

basically, a full agent training pipeline but compressed into 2.6B

base model → SFT → specialized teachers per domain (SFT + RLVR) → on-policy distillation back into one student → agentic RL

the two most interesting stages

→ MOPD: the student generates, each prompt routes to its domain teacher for token-level feedback. teachers branch from the same SFT checkpoint, so their signal stays close to the student's distribution

→ agentic RL: multi-turn GRPO inside real harnesses (OpenClaw, Hermes Agent), one sandbox per rollout, a proxy captures token-level trajectories while the harness stays a black box

this makes a 2.6B that beats much larger models on instruction following and tool use

SFT, distillation, RL, RL envs: exactly what we're covering in our Training Agents livestream series (next one coming soon!)

→ model: LiquidAI/LFM2.5-2.6B
→ blog: https://www.liquid.ai/blog/lfm2-5-2-6b
→ live series: https://www.youtube.com/playlist?list=PLo2EIpI_JMQvQZm-kVlz4wY1vWF0LBcf5

3 replies

liked 3 models 3 days ago

ManniX-ITA/Qwen3.6-27B-A3B-Coder

Text Generation • 27B • Updated 5 days ago • 607 • 5

LiquidAI/LFM2.5-2.6B-GGUF

Text Generation • 3B • Updated 1 day ago • 31.5k • 144

LiquidAI/LFM2.5-2.6B

Text Generation • 3B • Updated about 14 hours ago • 78k • 379

reacted to Banaxi-Tech's post with 🔥 4 days ago

Post

3221

We did an experiment, we wanted to see if AI is good enough to train models.
We used GPT 5.6 Sol Max for this because its one of the most powerful ones right now.
Our instructions were, it should write the training code, and start the training process and monitor it by itself.
We also gave it a link to BananaMind 2 Mini to get our architecture right.
The result: It worked, it made the working BananaMind 2 Nano, and even beat our previous MiniBananaMind v4 9M.
Its getting way easier to develop your own models now!