[PR #5098] [CLOSED] feat: support setting the KV cache quant type #11680

Closed
opened 2026-04-12 23:35:38 -05:00 by GiteaMirror · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/5098
Author: @sammcj
Created: 6/17/2024
Status: Closed

Base: mainHead: cacheconfig


📝 Commits (3)

  • 154d64b feat: support setting the KV cache quant type
  • 221f442 feat: support setting the KV cache quant type
  • 97e29ac Merge branch 'main' into cacheconfig

📊 Changes

5 files changed (+47 additions, -13 deletions)

View changed files

📝 api/types.go (+16 -12)
📝 cmd/cmd.go (+2 -0)
📝 envconfig/config.go (+14 -0)
📝 llm/server.go (+12 -1)
📝 parser/parser_test.go (+3 -0)

📄 Description

WIP

Testing adding configuration to allow setting the KV cache type re: #5091


  • Allow setting the KV cache type in the env and params.
  • Allow setting flashattention in params (as well as the existing env).

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/ollama/ollama/pull/5098 **Author:** [@sammcj](https://github.com/sammcj) **Created:** 6/17/2024 **Status:** ❌ Closed **Base:** `main` ← **Head:** `cacheconfig` --- ### 📝 Commits (3) - [`154d64b`](https://github.com/ollama/ollama/commit/154d64b2ad60e0f8d5e270df0f0370f862617da6) feat: support setting the KV cache quant type - [`221f442`](https://github.com/ollama/ollama/commit/221f442834cdb108da909438210883e121be5d5e) feat: support setting the KV cache quant type - [`97e29ac`](https://github.com/ollama/ollama/commit/97e29ac4fec8b607b856ee0255a777794174547d) Merge branch 'main' into cacheconfig ### 📊 Changes **5 files changed** (+47 additions, -13 deletions) <details> <summary>View changed files</summary> 📝 `api/types.go` (+16 -12) 📝 `cmd/cmd.go` (+2 -0) 📝 `envconfig/config.go` (+14 -0) 📝 `llm/server.go` (+12 -1) 📝 `parser/parser_test.go` (+3 -0) </details> ### 📄 Description WIP Testing adding configuration to allow setting the KV cache type re: #5091 --- - Allow setting the KV cache type in the env and params. - Allow setting flashattention in params (as well as the existing env). --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-04-12 23:35:38 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#11680