Requires a current llama.cpp / Ollama build with Gemma 4 (
gemma4_unified
) support.
Quality vs the full model
Held-out test split (10,514 examples),
description-only
, greedy. The bf16 column is the unquantized
transformers
model; Q8/Q4 are measured through Ollama.
metric
bf16
Q8_0
Q4_K_M
exact-match
0.714
0.697
0.682
micro-F1
0.756
0.732
0.718
macro-F1
0.538
0.500
0.429
easy exact
0.805
0.808
0.793
hard exact
0.644
0.611
0.595
Choosing a quant:
Q8_0 is effectively lossless. Q4_K_M costs only
1.5 points on exact/micro but **
7 points on macro-F1** — i.e. it degrades the
rare-CWE long tail
specifically. Use
Q8_0
if rare/long-tail CWEs matter; use
Q4_K_M
if size/speed dominate and you mostly see common CWEs.
Usage — Ollama
# build the model from the GGUF (see the Modelfile below)
ollama create cve-cwe-gemma4 -f Modelfile
# IMPORTANT: gemma4 runs with "thinking" ON by default, which wastes ~100-900# tokens per query for this single-label task. Disable it:
ollama run cve-cwe-gemma4
>>> /set nothink
>>> The update handler transmits user credentials over an unencrypted HTTP channel.
CWE-319
Via the API, pass
"think": false
(the
response
field is the clean answer):
curl -s http://localhost:11434/api/generate -d '{ "model": "cve-cwe-gemma4", "prompt": "SQL injection via the username parameter in the login form.", "stream": false, "think": false}'# -> {"response":"CWE-89", ...}
Modelfile
FROM ./cve-cwe-gemma4-12b-Q8_0.gguf
TEMPLATE """{{ if .System }}<|turn>system{{ .System }}<turn|>{{ end }}<|turn>user{{ .Prompt }}<turn|><|turn>model<|channel>thought<channel|>{{ .Response }}<turn|>"""
SYSTEM """You are a vulnerability analyst. Given a CVE description, reply with only the CWE ID(s) it maps to, comma-separated."""
PARAMETER temperature 0
PARAMETER stop "<turn|>"
PARAMETER stop "<|turn>"
Usage — llama.cpp
llama-cli -m cve-cwe-gemma4-12b-Q8_0.gguf --jinja \
-sys "You are a vulnerability analyst. Given a CVE description, reply with only the CWE ID(s) it maps to, comma-separated." \
-p "A heap-based buffer overflow occurs when parsing a crafted font file."
Limitations
Description-only triage aid; outputs are suggestions, not authoritative classifications. Keep a human in the loop for security-relevant decisions. Scope is MITRE View-1003. See the
base model card
for training details.
License
Apache-2.0, inherited from the Gemma 4 base model.
Runs of exploitintel cve-cwe-gemma4-12b-GGUF on huggingface.co
378
Total runs
9
24-hour runs
32
3-day runs
67
7-day runs
342
30-day runs
More Information About cve-cwe-gemma4-12b-GGUF huggingface.co Model
cve-cwe-gemma4-12b-GGUF huggingface.co is an AI model on huggingface.co that provides cve-cwe-gemma4-12b-GGUF's model effect (), which can be used instantly with this exploitintel cve-cwe-gemma4-12b-GGUF model. huggingface.co supports a free trial of the cve-cwe-gemma4-12b-GGUF model, and also provides paid use of the cve-cwe-gemma4-12b-GGUF. Support call cve-cwe-gemma4-12b-GGUF model through api, including Node.js, Python, http.
cve-cwe-gemma4-12b-GGUF huggingface.co is an online trial and call api platform, which integrates cve-cwe-gemma4-12b-GGUF's modeling effects, including api services, and provides a free online trial of cve-cwe-gemma4-12b-GGUF, you can try cve-cwe-gemma4-12b-GGUF online for free by clicking the link below.
exploitintel cve-cwe-gemma4-12b-GGUF online free url in huggingface.co:
cve-cwe-gemma4-12b-GGUF is an open source model from GitHub that offers a free installation service, and any user can find cve-cwe-gemma4-12b-GGUF on GitHub to install. At the same time, huggingface.co provides the effect of cve-cwe-gemma4-12b-GGUF install, users can directly use cve-cwe-gemma4-12b-GGUF installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
cve-cwe-gemma4-12b-GGUF install url in huggingface.co: