SyzygyResearch/Mach-1-Additive-35B
Updated • 2.51k • 122
Thivakar Jeyaseelan
Thiv
AI & ML interests
None yet
Recent Activity
liked a model 2 days ago
SyzygyResearch/Mach-1-Additive-35B liked a dataset 4 days ago
HuggingFaceFW/finewebOrganizations
None yet
liked a
model 2 days ago
liked a
dataset 4 days ago
reacted to mlabonne's post with ❤️ 4 days ago
Post
3067
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!
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 a
model 5 days ago
liked 4
models 6 days ago
reacted to sergiopaniego's post with 🤗 6 days ago
Post
2548
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
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 a
model 6 days ago
liked 2
models 7 days ago
reacted to Banaxi-Tech's post with 🔥 7 days ago
Post
3233
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!
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!
- 8 replies
liked 7
models 8 days ago
Comfy-Org/MiniMax-H3
Updated • 6.8M • 1.21k
BananaMind/BananaMind-2-Pro-Preview-Chat
Text Generation • 0.1B • Updated • 118 • 10
bench-labs/VoxelModel-v1
Text-to-3D • 40.2M • Updated • 1.57k • 17
SupraLabs/Supra2-100M-Instruct
Text Generation • 0.1B • Updated • 1.8k • 35
BananaMind/BananaMind-2-Pro-Preview
Text Generation • 0.2B • Updated • 276 • 22
bench-labs/PixelModel-v6
Text-to-Image • 0.2B • Updated • 6.12k • 18
bench-labs/PixelModel-v5
Text-to-Image • 0.2B • Updated • 5.95k • 31