Skip to content

Pull requests: beehive-lab/GPULlama3.java

New pull request New
11 Open 94 Closed
11 Open 94 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

OpenAI-compatible server (llama-tornado --server): /v1/chat/completions + /v1/completions, streaming, zero-dep
#135 opened Jul 14, 2026 by mikepapadim Member Loading…
2
On-device greedy sampling (-Dllama.deviceSample): GPU argmax, keep logits device-side (~500x less D2H)
#134 opened Jul 14, 2026 by mikepapadim Member Loading…
2
Add llama-bench-style benchmark: llama-tornado --bench (pp/tg/pg matrix, avg±stddev t/s, md/csv/json)
#133 opened Jul 14, 2026 by mikepapadim Member Loading…
@orionpapadakis
7
Fix Qwen3 FP16 garbage on CUDA backend: cross-workgroup race in RMS-norm reduction
#132 opened Jul 14, 2026 by mikepapadim Member Loading…
1
Hybrid CUDA library tasks for single-token decode: parity, not a win — measured findings + llama.logitsLib switch
#131 opened Jul 14, 2026 by mikepapadim Member Loading…
1
Static batched decode (LLaMA + Qwen3): B independent sequences per step, up to 41x aggregate throughput
#129 opened Jul 11, 2026 by mikepapadim Member Loading…
30
Fix IllegalArgumentException (negative ArrayList capacity) on prompts longer than max-tokens
#128 opened Jul 11, 2026 by mikepapadim Member Loading…
4
Add TornadoVM CUDA backend support and tensor-core (MMA) accelerated batch prefill (FP16 & Q8_0, Llama & Qwen3) enhancement New feature or request prefill-decode
#127 opened Jul 7, 2026 by mairooni Collaborator Loading…
@mairooni
Release 0.5.0
#126 opened Jun 25, 2026 by github-actions Bot Loading…
3 tasks
1
Add pure CUDA backend along the PTX engine
#124 opened Jun 20, 2026 by mikepapadim Member Loading…
11
Gemma 4 model support (CPU + GPU/TornadoVM, BF16 and Q8_0)
#120 opened Jun 7, 2026 by mikepapadim Member Loading…
3 tasks done
7
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.