automatically fit context on qwen3.5 2b and 4b

This commit is contained in:
2026-03-07 21:01:32 +01:00
parent 42038207fc
commit 8a074839b1

View File

@@ -599,7 +599,6 @@ models:
cmd: | cmd: |
/app/llama-server /app/llama-server
-hf unsloth/Qwen3.5-2B-GGUF:Q4_K_M -hf unsloth/Qwen3.5-2B-GGUF:Q4_K_M
--ctx-size 16384
--temp 0.6 --temp 0.6
--top-p 0.95 --top-p 0.95
--top-k 20 --top-k 20
@@ -613,7 +612,6 @@ models:
cmd: | cmd: |
/app/llama-server /app/llama-server
-hf unsloth/Qwen3.5-2B-GGUF:Q4_K_M -hf unsloth/Qwen3.5-2B-GGUF:Q4_K_M
--ctx-size 16384
--temp 0.6 --temp 0.6
--top-p 0.95 --top-p 0.95
--top-k 20 --top-k 20
@@ -627,7 +625,6 @@ models:
cmd: | cmd: |
/app/llama-server /app/llama-server
-hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M
--ctx-size 16384
--temp 0.6 --temp 0.6
--top-p 0.95 --top-p 0.95
--top-k 20 --top-k 20
@@ -641,7 +638,6 @@ models:
cmd: | cmd: |
/app/llama-server /app/llama-server
-hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M
--ctx-size 16384
--temp 0.6 --temp 0.6
--top-p 0.95 --top-p 0.95
--top-k 20 --top-k 20