Skip to content

Pull requests: AI-Hypercomputer/tpu-recipes

New pull request New
29 Open 241 Closed
29 Open 241 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

Feat(ironwood/vllm): add DeepSeek-R1 and DeepSeek-V3 serving recipes
#278 opened Aug 19, 2026 by inucool12 Loading…
3 tasks done
10
Add Qwen3.8-27B serving recipe for Ironwood (TPU7x)
#277 opened Aug 18, 2026 by R0rschach Loading…
7
Add Qwen3.8-27B serving recipe for Trillium (v6e)
#276 opened Aug 17, 2026 by R0rschach Loading…
4
Enable shard_exp_on_fsdp=True for Gemma 4 26B on Ironwood 4x4x4 recipe
#275 opened Aug 11, 2026 by csgoogle Contributor Loading…
Minor change in readme
#273 opened Aug 3, 2026 by Geeks-Sid Loading…
2
Add Z-Image-Turbo inference recipe for Trillium (TPU v6e)
#271 opened Jul 31, 2026 by csgoogle Contributor Loading…
1
Fix Qwen3-32B serving sequence caps and benchmark expectations
#268 opened Jul 17, 2026 by HazyRaven Loading…
1
Qwen3.6-27B recipe and README optimized for agentic coding #recipebot
#267 opened Jul 13, 2026 by jawadaminGOOG • Draft
Gemma4-31B-IT with MTP Speculative Decoding on Trillium TPU v6e-4
#261 opened Jun 18, 2026 by jawadaminGOOG • Draft
Feature/llama3.1 70b cluster toolkit
#254 opened May 27, 2026 by kvenkatachala333 • Draft
10
feat(llama3.1-8b): migrate MaxText training workload to gcluster
#253 opened May 25, 2026 by kvenkatachala333 • Draft
Add Qwen3 Coder 480B A35B instruct recipe on Ironwood #recipebot
#250 opened May 14, 2026 by alicherry-g • Draft
Update README.md Deepseek 256-chip Perf
#247 opened May 11, 2026 by cj-klein Contributor Loading…
Gemma4 vllm fixes
#227 opened Apr 20, 2026 by RobMulla Contributor Loading…
3
standardize docker build path and set stable mode for all Trillium recipes
#223 opened Apr 13, 2026 by sivanishwanthm Contributor Loading…
[GCS TPU7X NPI Recipe] Update GCS recipe to use GCS Direct and XLA Cache
#207 opened Mar 28, 2026 by lepan-google Contributor Loading…
1
#recipebot Add updated llama3.1-405b recipe on Ironwood
#179 opened Mar 9, 2026 by ankitkinra Collaborator • Draft
feat(recipe): Add Llama 3.1 70B 8k bf16 recipe for tpu7x-4x4x4
#107 opened Nov 21, 2025 by Krishna-hub-spec Loading…
1
add guide for SGLang-Jax on TPUs
#103 opened Oct 28, 2025 by JamesBrianD Loading…
10
feat: Add docker-compose and make based recipes
#97 opened Oct 13, 2025 by RobMulla Contributor Loading…
Docs: Update recipes for vLLM trillium with feedback.
#96 opened Oct 10, 2025 by RobMulla Contributor Loading…
1
Fix: Added vLLM CLI command to run benchmark in README.md
#89 opened Sep 17, 2025 by mzuo Loading…
Add a python gitignore
#78 opened May 2, 2025 by karan Collaborator Loading…
3
llama2 recipe: use py module, correctnesss fixes
#75 opened May 1, 2025 by karan Collaborator Loading…
3
Add MaxText Llama 3.1 70B training with GCS recipe
#56 opened Apr 9, 2025 by raymondzouu Collaborator Loading…
71
ProTip! Filter pull requests by the default branch with base:main.