Models
Datasets
Spaces
Buckets new
Docs
Enterprise
Pricing
- Website
- Community
- Solutions
Log In
Sign Up

Collections

Discover the best community collections!

Collections trending this week

Esper 4 is an agentic finetune focused on code, DevOps, architecture, and MLOps! Currently available for Qwen 3.6 and Gemma 4.

ValiantLabs/Qwen3.6-27B-Esper4

Image-Text-to-Text • 28B • Updated 25 days ago • 292 • 19
ValiantLabs/gemma-4-12B-it-Esper4

Image-Text-to-Text • 12B • Updated 25 days ago • 422 • 10

Next-generation Mixture-of-Experts (MoE) model optimized for advanced agentic coding and complex reasoning.

keras/qwen3_6_moe_35b_a3b

Text Generation • Updated 25 days ago • 439

prism-ml/Ternary-Bonsai-27B-gguf

Text Generation • 27B • Updated 22 days ago • 809k • • 1.18k

Microsoft's harrier is a decoder-only, multilingual text embedding model family using last-token pooling and L2 normalization

keras/bge_large_en

Updated 25 days ago • 31
keras/bge_base_v1.5_en

Updated 25 days ago • 32
keras/bge_small_v1.5_en

Updated 25 days ago • 33
keras/bge_large_v1.5_en

Updated 25 days ago • 39

This is a collection of Projects of RL Agents.

BJyotibrat/BanTRel-v1.0.0

Reinforcement Learning • Updated 26 days ago • 32 • 1
BJyotibrat/cartpole-dqn-converged

Reinforcement Learning • Updated Apr 5 • 4 • 1

CausalGame Reproduction (ICML 2026 #16331)

Artifacts for reproducing CausalGame: 14-setting causal-thinking benchmark for LLM agents. Bundle, results, oracle baselines.

Running

Repro - CausalGame: Benchmarking Causal Thinking of LLM Agents in Games

🎯

Explore game causal benchmarks with an interactive logbook
arinpt/causalgame-repro

Updated 25 days ago • 309
arinpt/causalgame-jobs

1.77 MB
arinpt/Repro-CausalGame-Benchmarking-Causal-Thinking-of-LLM-Agents-artifacts

2.1 MB

Highly efficient Mixture-of-Experts (MoE) large language model for advanced reasoning and text generation.

keras/qwen3_5_moe_35b_a3b_base

Text Generation • Updated 1 day ago • 434
keras/qwen3_5_moe_35b_a3b

Text Generation • Updated 1 day ago • 437

Harrier Embedding

Microsoft's harrier is a decoder-only, multilingual text embedding model family using last-token pooling and L2 normalization to generate Embeddings

keras/harrier_embedding_oss_270m

Updated 25 days ago • 32
keras/harrier_embedding_oss_0.6b

Updated 25 days ago • 22

Image Manipulation

Running on Zero

MCP

Featured

1.61k

FLUX.1 Kontext

⚡

1.61k

Kontext image editing on FLUX[dev]

opencerebral/Boris-250M

Text Generation • 0.3B • Updated 17 days ago • 107 • 1
opencerebral/Boris-250M-Instruct

Text Generation • 0.3B • Updated 17 days ago • 84
opencerebral/Boris-250M-Instruct-GGUF

Text Generation • 0.3B • Updated 17 days ago • 782 • 1
opencerebral/Boris-125M

Text Generation • 0.1B • Updated 4 days ago • 319

Esper 4 is an agentic finetune focused on code, DevOps, architecture, and MLOps! Currently available for Qwen 3.6 and Gemma 4.

ValiantLabs/Qwen3.6-27B-Esper4

Image-Text-to-Text • 28B • Updated 25 days ago • 292 • 19
ValiantLabs/gemma-4-12B-it-Esper4

Image-Text-to-Text • 12B • Updated 25 days ago • 422 • 10

CausalGame Reproduction (ICML 2026 #16331)

Artifacts for reproducing CausalGame: 14-setting causal-thinking benchmark for LLM agents. Bundle, results, oracle baselines.

Running

Repro - CausalGame: Benchmarking Causal Thinking of LLM Agents in Games

🎯

Explore game causal benchmarks with an interactive logbook
arinpt/causalgame-repro

Updated 25 days ago • 309
arinpt/causalgame-jobs

1.77 MB
arinpt/Repro-CausalGame-Benchmarking-Causal-Thinking-of-LLM-Agents-artifacts

2.1 MB

Next-generation Mixture-of-Experts (MoE) model optimized for advanced agentic coding and complex reasoning.

keras/qwen3_6_moe_35b_a3b

Text Generation • Updated 25 days ago • 439

Highly efficient Mixture-of-Experts (MoE) large language model for advanced reasoning and text generation.

keras/qwen3_5_moe_35b_a3b_base

Text Generation • Updated 1 day ago • 434
keras/qwen3_5_moe_35b_a3b

Text Generation • Updated 1 day ago • 437

prism-ml/Ternary-Bonsai-27B-gguf

Text Generation • 27B • Updated 22 days ago • 809k • • 1.18k

Harrier Embedding

Microsoft's harrier is a decoder-only, multilingual text embedding model family using last-token pooling and L2 normalization to generate Embeddings

keras/harrier_embedding_oss_270m

Updated 25 days ago • 32
keras/harrier_embedding_oss_0.6b

Updated 25 days ago • 22

Microsoft's harrier is a decoder-only, multilingual text embedding model family using last-token pooling and L2 normalization

keras/bge_large_en

Updated 25 days ago • 31
keras/bge_base_v1.5_en

Updated 25 days ago • 32
keras/bge_small_v1.5_en

Updated 25 days ago • 33
keras/bge_large_v1.5_en

Updated 25 days ago • 39

Image Manipulation

Running on Zero

MCP

Featured

1.61k

FLUX.1 Kontext

⚡

1.61k

Kontext image editing on FLUX[dev]

This is a collection of Projects of RL Agents.

BJyotibrat/BanTRel-v1.0.0

Reinforcement Learning • Updated 26 days ago • 32 • 1
BJyotibrat/cartpole-dqn-converged

Reinforcement Learning • Updated Apr 5 • 4 • 1

opencerebral/Boris-250M

Text Generation • 0.3B • Updated 17 days ago • 107 • 1
opencerebral/Boris-250M-Instruct

Text Generation • 0.3B • Updated 17 days ago • 84
opencerebral/Boris-250M-Instruct-GGUF

Text Generation • 0.3B • Updated 17 days ago • 782 • 1
opencerebral/Boris-125M

Text Generation • 0.1B • Updated 4 days ago • 319

Previous
1
...
639
640
641
642
643
...
22,358
Next

Company

TOS Privacy About Careers

Website

Models Datasets Spaces Pricing Docs