CVE-2026-102758 (GCVE-0-2026-102758)
Vulnerability from cvelistv5 – Published: 2026-09-29 17:26 – Updated: 2026-09-30 20:42
VLAI
EPSS
VEX
Summary
The `_nx_secure_x509_asn1_tlv_block_parse()` function parses ASN.1 TLV (tag-length-value) blocks out of DER-encoded data. It is the primitive underneath all X.509 certificate parsing in NetX Secure, and therefore runs on certificates supplied by a remote peer during the TLS handshake.
The function reads the one-byte ASN.1 tag from the caller's buffer *before* checking that the buffer holds at least one byte. When a caller passes a remaining length of zero, the guard correctly returns `NX_SECURE_X509_ASN1_LENGTH_TOO_LONG`, but the read has already happened one byte past the end of the buffer.
code:
nx_secure/src/nx_secure_x509_asn1_tlv_block_parse.c
```
UINT _nx_secure_x509_asn1_tlv_block_parse(const UCHAR *buffer, ULONG *buffer_length, USHORT *tlv_type,
USHORT *tlv_tag_class, ULONG *tlv_length,
const UCHAR **tlv_data, ULONG *header_length)
{
UINT current_index;
USHORT current_tag;
ULONG length;
ULONG length_bytes;
current_index = 0;
current_tag = buffer[current_index]; /* <-- read before the bounds check */
if (*buffer_length < 1)
{
return(NX_SECURE_X509_ASN1_LENGTH_TOO_LONG);
}
```
The remainder of the function is correctly ordered. The multi-byte length path is guarded by `length_bytes > 4 || length_bytes > *buffer_length` before its read loop, the decoded value is checked against `length > *buffer_length`, and the second single-byte length read follows its own `*buffer_length < 1` guard. The tag read is the only load placed ahead of its check.
Severity
7.5 (High)
SSVC
Exploitation: none
Automatable: yes
Technical Impact: partial
CISA Coordinator · CISA-ADP (v2.0.3)
Decision recorded 2026-09-30 20:42 UTC
CWE
- CWE-126 - Buffer Over-read
Assigner
References
1 reference
Impacted products
1 product
| Vendor | Product | Version | CPE status | |
|---|---|---|---|---|
| Eclipse Foundation | NetX Duo |
Affected:
0 , ≤ 6.5.1.202602
(custom)
Unaffected: 6.5.2.202603 |
guessed |
{
"containers": {
"adp": [
{
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 7.5,
"baseSeverity": "HIGH",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
}
},
{
"other": {
"content": {
"id": "CVE-2026-102758",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-30T20:42:14.455455Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-09-30T20:42:39.497Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"packageName": "NetX Duo",
"product": "NetX Duo",
"vendor": "Eclipse Foundation",
"versions": [
{
"lessThanOrEqual": "6.5.1.202602",
"status": "affected",
"version": "0",
"versionType": "custom"
},
{
"status": "unaffected",
"version": "6.5.2.202603"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "tinic"
}
],
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003eThe `_nx_secure_x509_asn1_tlv_block_parse()` function parses ASN.1 TLV (tag-length-value) blocks out of DER-encoded data. It is the primitive underneath all X.509 certificate parsing in NetX Secure, and therefore runs on certificates supplied by a remote peer during the TLS handshake.\u003c/p\u003e\u003cp\u003eThe function reads the one-byte ASN.1 tag from the caller\u0027s buffer *before* checking that the buffer holds at least one byte. When a caller passes a remaining length of zero, the guard correctly returns `NX_SECURE_X509_ASN1_LENGTH_TOO_LONG`, but the read has already happened one byte past the end of the buffer.\u003c/p\u003e\u003cp\u003ecode:\u003c/p\u003e\u003cp\u003enx_secure/src/nx_secure_x509_asn1_tlv_block_parse.c\u003c/p\u003e\u003cp\u003e```\u003c/p\u003e\u003cp\u003eUINT _nx_secure_x509_asn1_tlv_block_parse(const UCHAR *buffer, ULONG *buffer_length, USHORT *tlv_type,\u003c/p\u003e\u003ccode\u003e USHORT *tlv_tag_class, ULONG *tlv_length,\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e const UCHAR **tlv_data, ULONG *header_length)\u003c/code\u003e\u003cbr\u003e\u003cp\u003e{\u003c/p\u003e\u003cp\u003eUINT current_index;\u003c/p\u003e\u003cp\u003eUSHORT current_tag;\u003c/p\u003e\u003cp\u003eULONG length;\u003c/p\u003e\u003cp\u003eULONG length_bytes;\u003c/p\u003e\u003ccode\u003e current_index = 0;\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e current_tag = buffer[current_index]; /* \u0026lt;-- read before the bounds check */\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e if (*buffer_length \u0026lt; 1)\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e {\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e return(NX_SECURE_X509_ASN1_LENGTH_TOO_LONG);\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e }\u003c/code\u003e\u003cbr\u003e\u003cp\u003e```\u003c/p\u003e\u003cp\u003eThe remainder of the function is correctly ordered. The multi-byte length path is guarded by `length_bytes \u0026gt; 4 || length_bytes \u0026gt; *buffer_length` before its read loop, the decoded value is checked against `length \u0026gt; *buffer_length`, and the second single-byte length read follows its own `*buffer_length \u0026lt; 1` guard. The tag read is the only load placed ahead of its check.\u003c/p\u003e"
}
],
"value": "The `_nx_secure_x509_asn1_tlv_block_parse()` function parses ASN.1 TLV (tag-length-value) blocks out of DER-encoded data. It is the primitive underneath all X.509 certificate parsing in NetX Secure, and therefore runs on certificates supplied by a remote peer during the TLS handshake.\n\n\n\nThe function reads the one-byte ASN.1 tag from the caller\u0027s buffer *before* checking that the buffer holds at least one byte. When a caller passes a remaining length of zero, the guard correctly returns `NX_SECURE_X509_ASN1_LENGTH_TOO_LONG`, but the read has already happened one byte past the end of the buffer.\n\n\n\ncode:\n\n\n\nnx_secure/src/nx_secure_x509_asn1_tlv_block_parse.c\n\n\n\n```\n\n\n\nUINT _nx_secure_x509_asn1_tlv_block_parse(const UCHAR *buffer, ULONG *buffer_length, USHORT *tlv_type,\n\n USHORT *tlv_tag_class, ULONG *tlv_length,\n const UCHAR **tlv_data, ULONG *header_length)\n\n\n{\n\n\n\nUINT current_index;\n\n\n\nUSHORT current_tag;\n\n\n\nULONG length;\n\n\n\nULONG length_bytes;\n\n current_index = 0;\n current_tag = buffer[current_index]; /* \u003c-- read before the bounds check */\n if (*buffer_length \u003c 1)\n {\n return(NX_SECURE_X509_ASN1_LENGTH_TOO_LONG);\n }\n\n\n```\n\n\n\nThe remainder of the function is correctly ordered. The multi-byte length path is guarded by `length_bytes \u003e 4 || length_bytes \u003e *buffer_length` before its read loop, the decoded value is checked against `length \u003e *buffer_length`, and the second single-byte length read follows its own `*buffer_length \u003c 1` guard. The tag read is the only load placed ahead of its check."
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-126",
"description": "CWE-126 Buffer Over-read",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-09-29T17:26:44.424Z",
"orgId": "e51fbebd-6053-4e49-959f-1b94eeb69a2c",
"shortName": "eclipse"
},
"references": [
{
"url": "https://github.com/eclipse-threadx/netxduo/security/advisories/GHSA-f95g-xc4w-mrcr"
}
],
"source": {
"discovery": "UNKNOWN"
},
"x_generator": {
"engine": "Vulnogram 1.0.5"
}
}
},
"cveMetadata": {
"assignerOrgId": "e51fbebd-6053-4e49-959f-1b94eeb69a2c",
"assignerShortName": "eclipse",
"cveId": "CVE-2026-102758",
"datePublished": "2026-09-29T17:26:44.424Z",
"dateReserved": "2026-09-29T16:21:58.057Z",
"dateUpdated": "2026-09-30T20:42:39.497Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"epss": {
"cve": "CVE-2026-102758",
"date": "2026-11-13",
"epss": "0.00173",
"percentile": "0.06848"
},
"nvd": {
"cve": {
"affected": [
{
"affectedData": [
{
"defaultStatus": "unaffected",
"packageName": "NetX Duo",
"product": "NetX Duo",
"vendor": "Eclipse Foundation",
"versions": [
{
"lessThanOrEqual": "6.5.1.202602",
"status": "affected",
"version": "0",
"versionType": "custom"
},
{
"status": "unaffected",
"version": "6.5.2.202603"
}
]
}
],
"source": "emo@eclipse.org"
}
],
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "The `_nx_secure_x509_asn1_tlv_block_parse()` function parses ASN.1 TLV (tag-length-value) blocks out of DER-encoded data. It is the primitive underneath all X.509 certificate parsing in NetX Secure, and therefore runs on certificates supplied by a remote peer during the TLS handshake.\n\n\n\nThe function reads the one-byte ASN.1 tag from the caller\u0027s buffer *before* checking that the buffer holds at least one byte. When a caller passes a remaining length of zero, the guard correctly returns `NX_SECURE_X509_ASN1_LENGTH_TOO_LONG`, but the read has already happened one byte past the end of the buffer.\n\n\n\ncode:\n\n\n\nnx_secure/src/nx_secure_x509_asn1_tlv_block_parse.c\n\n\n\n```\n\n\n\nUINT _nx_secure_x509_asn1_tlv_block_parse(const UCHAR *buffer, ULONG *buffer_length, USHORT *tlv_type,\n\n USHORT *tlv_tag_class, ULONG *tlv_length,\n const UCHAR **tlv_data, ULONG *header_length)\n\n\n{\n\n\n\nUINT current_index;\n\n\n\nUSHORT current_tag;\n\n\n\nULONG length;\n\n\n\nULONG length_bytes;\n\n current_index = 0;\n current_tag = buffer[current_index]; /* \u003c-- read before the bounds check */\n if (*buffer_length \u003c 1)\n {\n return(NX_SECURE_X509_ASN1_LENGTH_TOO_LONG);\n }\n\n\n```\n\n\n\nThe remainder of the function is correctly ordered. The multi-byte length path is guarded by `length_bytes \u003e 4 || length_bytes \u003e *buffer_length` before its read loop, the decoded value is checked against `length \u003e *buffer_length`, and the second single-byte length read follows its own `*buffer_length \u003c 1` guard. The tag read is the only load placed ahead of its check."
}
],
"id": "CVE-2026-102758",
"lastModified": "2026-09-30T21:17:05.347",
"metrics": {
"cvssMetricV31": [
{
"cvssData": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 7.5,
"baseSeverity": "HIGH",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"exploitabilityScore": 3.9,
"impactScore": 3.6,
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"type": "Secondary"
}
],
"ssvcV203": [
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"ssvcData": {
"id": "CVE-2026-102758",
"options": [
{
"exploitation": "none"
},
{
"automatable": "yes"
},
{
"technicalImpact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-30T20:42:14.455455Z",
"version": "2.0.3"
}
}
]
},
"published": "2026-09-29T18:17:13.047",
"references": [
{
"source": "emo@eclipse.org",
"url": "https://github.com/eclipse-threadx/netxduo/security/advisories/GHSA-f95g-xc4w-mrcr"
}
],
"sourceIdentifier": "emo@eclipse.org",
"vulnStatus": "Awaiting Analysis",
"weaknesses": [
{
"description": [
{
"lang": "en",
"value": "CWE-126"
}
],
"source": "emo@eclipse.org",
"type": "Secondary"
}
]
}
},
"vulnrichment": {
"containers": {
"adp": [
{
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 7.5,
"baseSeverity": "HIGH",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"version": "3.1"
}
},
{
"other": {
"content": {
"id": "CVE-2026-102758",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-09-30T20:42:14.455455Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-09-30T20:42:35.929Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"packageName": "NetX Duo",
"product": "NetX Duo",
"vendor": "Eclipse Foundation",
"versions": [
{
"lessThanOrEqual": "6.5.1.202602",
"status": "affected",
"version": "0",
"versionType": "custom"
},
{
"status": "unaffected",
"version": "6.5.2.202603"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "tinic"
}
],
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003eThe `_nx_secure_x509_asn1_tlv_block_parse()` function parses ASN.1 TLV (tag-length-value) blocks out of DER-encoded data. It is the primitive underneath all X.509 certificate parsing in NetX Secure, and therefore runs on certificates supplied by a remote peer during the TLS handshake.\u003c/p\u003e\u003cp\u003eThe function reads the one-byte ASN.1 tag from the caller\u0027s buffer *before* checking that the buffer holds at least one byte. When a caller passes a remaining length of zero, the guard correctly returns `NX_SECURE_X509_ASN1_LENGTH_TOO_LONG`, but the read has already happened one byte past the end of the buffer.\u003c/p\u003e\u003cp\u003ecode:\u003c/p\u003e\u003cp\u003enx_secure/src/nx_secure_x509_asn1_tlv_block_parse.c\u003c/p\u003e\u003cp\u003e```\u003c/p\u003e\u003cp\u003eUINT _nx_secure_x509_asn1_tlv_block_parse(const UCHAR *buffer, ULONG *buffer_length, USHORT *tlv_type,\u003c/p\u003e\u003ccode\u003e USHORT *tlv_tag_class, ULONG *tlv_length,\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e const UCHAR **tlv_data, ULONG *header_length)\u003c/code\u003e\u003cbr\u003e\u003cp\u003e{\u003c/p\u003e\u003cp\u003eUINT current_index;\u003c/p\u003e\u003cp\u003eUSHORT current_tag;\u003c/p\u003e\u003cp\u003eULONG length;\u003c/p\u003e\u003cp\u003eULONG length_bytes;\u003c/p\u003e\u003ccode\u003e current_index = 0;\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e current_tag = buffer[current_index]; /* \u0026lt;-- read before the bounds check */\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e if (*buffer_length \u0026lt; 1)\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e {\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e return(NX_SECURE_X509_ASN1_LENGTH_TOO_LONG);\u003c/code\u003e\u003cbr\u003e\u003ccode\u003e }\u003c/code\u003e\u003cbr\u003e\u003cp\u003e```\u003c/p\u003e\u003cp\u003eThe remainder of the function is correctly ordered. The multi-byte length path is guarded by `length_bytes \u0026gt; 4 || length_bytes \u0026gt; *buffer_length` before its read loop, the decoded value is checked against `length \u0026gt; *buffer_length`, and the second single-byte length read follows its own `*buffer_length \u0026lt; 1` guard. The tag read is the only load placed ahead of its check.\u003c/p\u003e"
}
],
"value": "The `_nx_secure_x509_asn1_tlv_block_parse()` function parses ASN.1 TLV (tag-length-value) blocks out of DER-encoded data. It is the primitive underneath all X.509 certificate parsing in NetX Secure, and therefore runs on certificates supplied by a remote peer during the TLS handshake.\n\n\n\nThe function reads the one-byte ASN.1 tag from the caller\u0027s buffer *before* checking that the buffer holds at least one byte. When a caller passes a remaining length of zero, the guard correctly returns `NX_SECURE_X509_ASN1_LENGTH_TOO_LONG`, but the read has already happened one byte past the end of the buffer.\n\n\n\ncode:\n\n\n\nnx_secure/src/nx_secure_x509_asn1_tlv_block_parse.c\n\n\n\n```\n\n\n\nUINT _nx_secure_x509_asn1_tlv_block_parse(const UCHAR *buffer, ULONG *buffer_length, USHORT *tlv_type,\n\n USHORT *tlv_tag_class, ULONG *tlv_length,\n const UCHAR **tlv_data, ULONG *header_length)\n\n\n{\n\n\n\nUINT current_index;\n\n\n\nUSHORT current_tag;\n\n\n\nULONG length;\n\n\n\nULONG length_bytes;\n\n current_index = 0;\n current_tag = buffer[current_index]; /* \u003c-- read before the bounds check */\n if (*buffer_length \u003c 1)\n {\n return(NX_SECURE_X509_ASN1_LENGTH_TOO_LONG);\n }\n\n\n```\n\n\n\nThe remainder of the function is correctly ordered. The multi-byte length path is guarded by `length_bytes \u003e 4 || length_bytes \u003e *buffer_length` before its read loop, the decoded value is checked against `length \u003e *buffer_length`, and the second single-byte length read follows its own `*buffer_length \u003c 1` guard. The tag read is the only load placed ahead of its check."
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-126",
"description": "CWE-126 Buffer Over-read",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-09-29T17:26:44.424Z",
"orgId": "e51fbebd-6053-4e49-959f-1b94eeb69a2c",
"shortName": "eclipse"
},
"references": [
{
"url": "https://github.com/eclipse-threadx/netxduo/security/advisories/GHSA-f95g-xc4w-mrcr"
}
],
"source": {
"discovery": "UNKNOWN"
},
"x_generator": {
"engine": "Vulnogram 1.0.5"
}
}
},
"cveMetadata": {
"assignerOrgId": "e51fbebd-6053-4e49-959f-1b94eeb69a2c",
"assignerShortName": "eclipse",
"cveId": "CVE-2026-102758",
"datePublished": "2026-09-29T17:26:44.424Z",
"dateReserved": "2026-09-29T16:21:58.057Z",
"dateUpdated": "2026-09-30T20:42:39.497Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
}
}
Loading…
Loading…
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
Loading…
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Loading…
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.