SqueezeAILab

SqueezeAI is part of Berkeley AI Research Lab at UC Berkeley focused on AI Systems research.

Popular repositoriesLoading

LLMCompilerLLMCompilerPublic
[ICML 2024] LLMCompiler: An LLM Compiler for Parallel Function Calling
Python 1.8k 127
SqueezeLLMSqueezeLLMPublic
[ICML 2024] SqueezeLLM: Dense-and-Sparse Quantization
Python 713 49
TinyAgentTinyAgentPublic
[EMNLP 2024 Demo] TinyAgent: Function Calling at the Edge!
Python 467 71
KVQuantKVQuantPublic
[NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization
Python 404 38
LLM2LLMLLM2LLMPublic
[ACL 2024] LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
Python 193 16
SqueezedAttentionSqueezedAttentionPublic
[ACL 2025] Squeezed Attention: Accelerating Long Prompt LLM Inference
Python 56 8

Repositories

Showing 10 of 16 repositories

Arbitrage Public
Arbitrage: Efficient Reasoning via Advantage-Aware Speculation
SqueezeAILab/Arbitrage’s past year of commit activity
Python 30 1 0 UpdatedDec 10, 2025
MultipoleAttention Public
[NeurIPS 2025] Multipole Attention for Efficient Long Context Reasoning
SqueezeAILab/MultipoleAttention’s past year of commit activity
Python 200 2 0 UpdatedDec 5, 2025
CDLM Public
CDLM: Consistency Diffusion Language Models for Faster Sampling
SqueezeAILab/CDLM’s past year of commit activity
Python 20MIT0 0 0 UpdatedNov 25, 2025
plan-and-act Public
[ICML 2025] Improving Planning of Agents for Long-Horizon Tasks
SqueezeAILab/plan-and-act’s past year of commit activity
Python 22MIT 3 0 0 UpdatedOct 2, 2025
sciml-agent Public
SciMLAgents: Write the Solver, Not the Solution
SqueezeAILab/sciml-agent’s past year of commit activity
4MIT0 0 0 UpdatedSep 15, 2025
reward-under-attack Public
SqueezeAILab/reward-under-attack’s past year of commit activity
00 0 0 UpdatedJul 26, 2025
ETS Public
ETS: Efficient Tree Search for Inference-Time Scaling
SqueezeAILab/ETS’s past year of commit activity
Python 9 1 1 0 UpdatedFeb 28, 2025
QuantSpec Public
SqueezeAILab/QuantSpec’s past year of commit activity
70 2 0 UpdatedFeb 22, 2025
SqueezedAttention Public
[ACL 2025] Squeezed Attention: Accelerating Long Prompt LLM Inference
SqueezeAILab/SqueezedAttention’s past year of commit activity
Python 56 8 4 0 UpdatedNov 20, 2024
Tool2Vec Public
Efficient and Scalable Estimation of Tool Representations in Vector Space
SqueezeAILab/Tool2Vec’s past year of commit activity
Python 29MIT 4 3 0 UpdatedSep 5, 2024