From 9f0183e34c2b339c15de22ca2d9f7ccf30e36de0 Mon Sep 17 00:00:00 2001
From: Ronald Tse <ronald.tse@ribose.com>
Date: Tue, 1 Sep 2026 06:51:48 +0200
Subject: [PATCH] =?UTF-8?q?feat(eval):=20GLM=5FWORKERS=20env=20knob=20?=
 =?UTF-8?q?=E2=80=94=208=20default,=20lower=20for=20loaded=20endpoints?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

glm-4.7-flash currently sits behind sustained 429s (code 1305);
hammering it with 8 concurrent workers burns the 5-retry budget into
empty rows. The knob lets overloaded models run politely; empties
remain self-healing on resume (#65).
---
 eval_sadeed_glm.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/eval_sadeed_glm.py b/eval_sadeed_glm.py
index d9ff0a7..003d727 100644
--- a/eval_sadeed_glm.py
+++ b/eval_sadeed_glm.py
@@ -132,7 +132,8 @@ def work(i: int) -> tuple[int, str]:
             return i, call(session, key, inputs[i])
 
         n_written = 0
-        with ThreadPoolExecutor(max_workers=8) as ex:
+        workers = int(os.environ.get("GLM_WORKERS", "8"))
+        with ThreadPoolExecutor(max_workers=workers) as ex:
             futures = {ex.submit(work, i): i for i in todo}
             for fut in as_completed(futures):
                 i, pred = fut.result()