[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[elpa] externals/llm 32fadc0fd8 2/2: Fix how we look for ollama models i
From: |
ELPA Syncer |
Subject: |
[elpa] externals/llm 32fadc0fd8 2/2: Fix how we look for ollama models in integration tests (#80) |
Date: |
Sat, 7 Sep 2024 00:58:18 -0400 (EDT) |
branch: externals/llm
commit 32fadc0fd87c06d84e545a02e279eb5f9f6f4bd5
Author: Andrew Hyatt <ahyatt@gmail.com>
Commit: GitHub <noreply@github.com>
Fix how we look for ollama models in integration tests (#80)
Ollama still doesn't perform very well in integration tests in general
due to most models having poor instruction following.
---
llm-integration-test.el | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/llm-integration-test.el b/llm-integration-test.el
index cac5b981e8..78968e4076 100644
--- a/llm-integration-test.el
+++ b/llm-integration-test.el
@@ -102,7 +102,7 @@
(when (getenv "VERTEX_PROJECT")
(require 'llm-vertex)
(push (make-llm-vertex :project (getenv "VERTEX_PROJECT")) providers))
- (when (getenv "OLLAMA_MODELS")
+ (when (getenv "OLLAMA_CHAT_MODELS")
(require 'llm-ollama)
;; This variable is a list of models to test.
(dolist (model (split-string (getenv "OLLAMA_CHAT_MODELS") ", "))