Models
MoE models from 35B to 398B parameters. Run models larger than your VRAM - expert offloading handles the rest. Every available model's output is verified against llama.cpp.
Each model ships as a single .lmpack. The quantization it is built from varies by model and is listed on each model page.
Available now
Benchmarked and verified on real hardwarelean-agent-middle
Middleweight class · v1.0 · Qwen 3.5 base
Advanced agent - complex orchestration, long-context workflows
Total params
122B
Active per token
10B
Base model
Qwen3.5-122B-A10B
Architecture
GDN hybrid MoE
Min VRAM
24 GB
Tool calling
Non-streaming
122B total, 10B active per token. A 75 GB model that runs on a single 24 GB GPU through expert offloading - massive knowledge with efficient per-token compute.
Not downloadable yet - checkout is being finalized
lean-coder-welter
Welterweight class · v1.0 · Qwen 3-Coder-Next base
Code generation - debugging, refactoring, code review
Total params
80B
Active per token
3B
Base model
Qwen3-Coder-Next
Architecture
MoE (512 experts)
Min VRAM
12 GB
Tool calling
Non-streaming
80B total with 512 experts, only 3B active per token. A 49.7 GB model that runs on 12 GB VRAM. Tuned for code generation, debugging, and software engineering.
$ lean pull lean-coder-welterlean-agent-light
Lightweight class · v1.0 · Qwen 3.5 base
General-purpose agent - tool calling, structured output, multi-step reasoning
Total params
35B
Active per token
3B
Base model
Qwen3.5-35B-A3B
Architecture
GDN hybrid MoE
Min VRAM
12 GB
Tool calling
Non-streaming
The entry point. A 22 GB model that runs on 12 GB VRAM - expert offloading handles the rest. Qwen3.5 GDN hybrid architecture surpasses last-gen models many times its size.
$ lean pull lean-agent-lightNot yet released
Validation pendingThese haven't been benchmarked or verified on our hardware yet. Performance and hardware requirements are targets, not measured results. Download sizes are measured wherever the model is already packed. No download or purchase until they clear the same verification the available models passed.
lean-agent-cruiser
Cruiserweight class · v1.0 · DeepSeek V4-Flash base
Million-token context - frontier long-context work on a single GPU
Total params
284B
Active per token
13B
Base model
DeepSeek-V4-Flash
Architecture
Hybrid CSA/HCA attention MoE
Target VRAM
24 GB
284B total, 13B active per token. DeepSeek's hybrid attention pairs Compressed Sparse Attention with Heavily Compressed Attention to hold a 1,048,576-token context, and Manifold-Constrained Hyper-Connections stabilise signal propagation across its 43 layers.
lean-think-heavy
Heavyweight class · v1.0 · Arcee Trinity Large-Thinking base
Extended reasoning - chain-of-thought, agentic tasks, deep analysis
Total params
398B
Active per token
~13B
Base model
Arcee Trinity-Large-Thinking
Architecture
afmoe (SWA + global)
Target VRAM
48 GB
398B total, ~13B active per token. Interleaved sliding-window + global attention with chain-of-thought and agentic RL post-training. The largest model in the catalog.
lean-reason-heavy
Heavyweight class · v1.0 · Qwen 3.5 base
Near-frontier - deep reasoning, complex analysis, research
Total params
397B
Active per token
17B
Base model
Qwen3.5-397B-A17B
Architecture
GDN hybrid MoE
Target VRAM
48 GB
397B total, 17B active per token. Near-frontier reasoning from a massive expert pool, running entirely on your hardware.
How offloading works
MoE models only activate a fraction of their parameters per token. The lean runtime keeps the hot path in VRAM and transparently pages in the rest from RAM and NVMe as needed.
The .lmpack format is designed for this workload. Combined with speculative prefetching and profile-guided preloading, it delivers interactive speeds on hardware that would otherwise be far too small.
Get started
$ curl -sSf https://leanmodels.ai/install.sh | sh
$ lean pull lean-agent-light
$ lean run lean-agent-lightSingle binary, 15 MB. No Python, no Docker, no cloud dependency.