openclaw/hermes-agent
1
0
Fork 0
Code Issues Pull Requests Actions 9 Packages Projects Releases Wiki Activity

fix(minimax): drop stale ≤204,800 cache entries for MiniMax-M3 (#36726)

Browse Source 
M3 is 1M context, but pre-catalog builds resolved it via the generic
'minimax' catch-all (204,800) and persisted that to the context-length
cache. Step 1 of get_model_context_length returned the cached value
directly before reaching the 'minimax-m3' (1M) catalog entry, so users
who first probed M3 on an older build were stuck at 204K forever (e.g.
/new in the Telegram gateway showing 'Context: 204K tokens (detected)').

Mirror the existing Kimi/Codex stale-cache guards: when a cached entry
for a minimax-m3 slug is <= 204,800, drop it and re-resolve. M2.x slugs
(correctly 204,800) are untouched since they don't match the M3 name.
This commit is contained in:
Teknium
2026-06-01 14:59:07 -07:00
committed by GitHub
parent b9646276fd
commit 1ffa22ee6b
2 changed files with 91 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
agent/model_metadata.py
View File

@ -1128,6 +1128,18 @@ def _model_name_suggests_kimi(model: str) -> bool:
return lower.startswith("kimi") or "moonshot" in lower
def _model_name_suggests_minimax_m3(model: str) -> bool:
"""Return True if the model name looks like MiniMax M3.
Catches ``MiniMax-M3``, ``minimax/minimax-m3``, and similar variants
across surfaces (native MiniMax-M3, OpenRouter/Nous minimax/minimax-m3).
Used as a guard against stale cache entries seeded by pre-catalog builds
that resolved M3 via the generic ``minimax`` catch-all (204,800) before
the ``minimax-m3`` (1M) entry existed in DEFAULT_CONTEXT_LENGTHS.
"""
return "minimax-m3" in model.lower()
def _query_local_context_length(model: str, base_url: str, api_key: str = "") -> Optional[int]:
"""Query a local server for the model's context length."""
import httpx
@ -1539,6 +1551,19 @@ def get_model_context_length(
model, base_url, f"{cached:,}",
)
_invalidate_cached_context_length(model, base_url)
# Invalidate stale ≤204,800 cache entries for MiniMax-M3. Pre-catalog
# builds resolved M3 via the generic ``minimax`` catch-all (204,800)
# and persisted it before the ``minimax-m3`` (1M) entry existed; that
# stale value would otherwise stick forever here at step 1. M3 is 1M,
# so any sub-256K cached value for an M3 slug is a leftover — drop it
# and fall through to the hardcoded default.
elif cached <= 204_800 and _model_name_suggests_minimax_m3(model):
logger.info(
"Dropping stale MiniMax-M3 cache entry %s@%s -> %s (pre-catalog value); "
"re-resolving via hardcoded defaults",
model, base_url, f"{cached:,}",
)
_invalidate_cached_context_length(model, base_url)
# Nous Portal: the portal /v1/models endpoint is authoritative.
# Bypass the persistent cache so step 5b can always reconcile
# against it — this corrects pre-fix entries seeded from the