WUMBOLABS / EVALUATIONS CANONICAL MODEL EVALUATION

/evaluations/neohorse-1-9b

NeoHorse-1-9B

Published evaluation evidence and testing history. Results remain attributed to their original configurations and methods.

latest evidence 2026-09-25

Overview

TokenRhythm · 2 tested configurations · 3 evidence events · newest evidence 2026-09-25.

Contents: Findings · Event results · Tested configurations · Chronology · Technical provenance

Findings

Each finding is attributed to the event and tested configuration that measured it; exact values, failures and limitations are in the event results. A historical event's recorded status describes that event's own sections at its date and is not a verdict on newer sections.

  • Model classification: READY_WITH_GUARDRAILS - attributed to the 2026-09-25 event, configuration llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070.
  • Recommended configuration: llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K) (current-alternate): event result.
  • Latest evidence: 2026-09-25: event result
  • 2026-09-25 · llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070: classification READY_WITH_GUARDRAILS. Event result
  • 2026-09-25 · llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070: classification READY_WITH_GUARDRAILS. Event result
  • 2026-09-25 · llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070: classification READY_WITH_GUARDRAILS. Event result

Event results

Newest first. Each event is one independent testing/publication event with its full report linked once at the top of its section; scores, failures and limitations stay visible and raw section-provenance JSON stays collapsed.

2026-09-25: First fresh-model WELP campaign under the model-agentic snapshot: full Model science of the publisher-default Reasoning On profile; case C measured (Reasoning Off sibling required)

WELP Fresh-Model Characterization: NeoHorse-1-9B (Reasoning On, publisher default) · config llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070 · maturity: CURRENT_WELP · status: READY_WITH_GUARDRAILS

Configuration: llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K).

Full report (REPORT.md)

WELP outcome

  • Outcome: COMPLETE_PASS
  • Classification: not recorded

Publication state: published; canonical evidence linked above

Quality and capabilities

  • Constrained result: not recorded
Guardrails and limitations

Reliability: not recorded

LocalMaxxing

  • Status: SUBMITTED
  • Submission reference: not recorded
  • verifiedRun: null (not claimed)

canonical practical profile Q8_0 llama.cpp; the benchmark workload is reasoning-state invariant

Tested condition: llama.cpp b10999 · RTX 5070.

Full tested identity, runtime and context details (exact hashes, versions and configuration)

Identity

FieldValue
Profile identityneohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-on
ModelNeoHorse-1-9B
Producernot recorded
Tested artifactnot recorded
Precisionnot recorded
Campaignneohorse-1-9b-rtx5070-welp-reasoning-on-2026-09-25
Record date2026-09-26

Runtime and hardware

FieldValue
Enginellama.cpp b10999
Runtime tagb10999
Runtime commitb04d4e567cd2fb8d2ded6e17d38dbbcfafe29063
Runtime buildbuild-cuda-sm120
HardwareNVIDIA GeForce RTX 5070 12 GB

Context profile

FieldValue
Practical defaultnot recorded
Guarded contextnot recorded
Native model-card maximumnot recorded
Model-card envelope completenot recorded
Native maximum dispositionnot recorded

2026-09-25: Full WELP characterization of the Reasoning Off deployment profile: independent setup/calibration, same frozen task set, dual-profile completion of the case-C model

WELP Reasoning-Profile Characterization: NeoHorse-1-9B (Reasoning Off, supported alternate) · config llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070 · maturity: CURRENT_WELP · status: READY_WITH_GUARDRAILS

Configuration: llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K).

Full report (REPORT.md)

WELP outcome

  • Outcome: COMPLETE_PASS
  • Classification: not recorded

Publication state: published; canonical evidence linked above

Quality and capabilities

  • Constrained result: not recorded
Guardrails and limitations

Reliability: not recorded

LocalMaxxing

  • Status: SUBMITTED
  • Submission reference: not recorded
  • verifiedRun: null (not claimed)

canonical practical profile Q8_0 llama.cpp; the benchmark workload is reasoning-state invariant

Tested condition: llama.cpp b10999 · RTX 5070.

Full tested identity, runtime and context details (exact hashes, versions and configuration)

Identity

FieldValue
Profile identityneohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-off
ModelNeoHorse-1-9B
Producernot recorded
Tested artifactnot recorded
Precisionnot recorded
Campaignneohorse-1-9b-rtx5070-welp-reasoning-off-2026-09-25
Record date2026-09-26

Runtime and hardware

FieldValue
Enginellama.cpp b10999
Runtime tagb10999
Runtime commitb04d4e567cd2fb8d2ded6e17d38dbbcfafe29063
Runtime buildbuild-cuda-sm120
HardwareNVIDIA GeForce RTX 5070 12 GB

Context profile

FieldValue
Practical defaultnot recorded
Guarded contextnot recorded
Native model-card maximumnot recorded
Model-card envelope completenot recorded
Native maximum dispositionnot recorded

2026-09-25: First fresh-model WELP Agentic event: dual-adapter qualification, three required task classes with two scored (1 PASS / 1 FAIL) and the system class INTEGRATION_BLOCKED by a deterministic frozen-harness defect

WELP Agentic: NeoHorse-1-9B (publisher-default profile, native tool calls) · config llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070 · maturity: CURRENT_WELP · status: READY_WITH_GUARDRAILS

Configuration: llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K).

Full report (REPORT.md)

WELP outcome

  • Outcome: COMPLETE_PASS
  • Classification: not recorded

Publication state: published; canonical evidence linked above

Quality and capabilities

  • Constrained result: not recorded
Guardrails and limitations

Reliability: not recorded

LocalMaxxing

  • Status: SUBMITTED
  • Submission reference: not recorded
  • verifiedRun: null (not claimed)

canonical practical profile Q8_0 llama.cpp; the benchmark workload is reasoning-state invariant

Tested condition: llama.cpp b10999 · RTX 5070.

Full tested identity, runtime and context details (exact hashes, versions and configuration)

Identity

FieldValue
Profile identityneohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-on
ModelNeoHorse-1-9B
Producernot recorded
Tested artifactnot recorded
Precisionnot recorded
Campaignneohorse-1-9b-rtx5070-welp-agentic-2026-09-25
Record date2026-09-26

Runtime and hardware

FieldValue
Enginellama.cpp b10999
Runtime tagb10999
Runtime commitb04d4e567cd2fb8d2ded6e17d38dbbcfafe29063
Runtime buildbuild-cuda-sm120
HardwareNVIDIA GeForce RTX 5070 12 GB

Context profile

FieldValue
Practical defaultnot recorded
Guarded contextnot recorded
Native model-card maximumnot recorded
Model-card envelope completenot recorded
Native maximum dispositionnot recorded

Tested configurations

Exact profile identities, sampling conditions and provenance are in each configuration's details below.

ConfigurationArtifact / precisionRuntimeHardwareContext / reasoningEvents
llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070Q8_0 official publisher GGUFllama.cpp b10999RTX 5070not recorded2 (2026-09-25, 2026-09-25)
llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070Q8_0 official publisher GGUFllama.cpp b10999RTX 5070not recorded1 (2026-09-25)
Configuration details: llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070
  • Full profile name: llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K).
  • Profile identity: neohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-on.
  • Status: current tested surface; not the recommended profile.
  • Artifact (as recorded): TokenRhythm/NeoHorse-1-9B-GGUF@ddcb4c93 :: NeoHorse-1-9B-Q8_0.gguf (sha256 519869730bda973ec50bb3ac42cd874569e5f3a57c3e0d5d23a2b5040e93310f)
  • Runtime (as recorded): llama.cpp b10999.
  • Hardware (as recorded): NVIDIA GeForce RTX 5070 12 GB.
  • Profile Metadata

Events on this configuration:

Configuration details: llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070
  • Full profile name: llama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K).
  • Profile identity: neohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-off.
  • Status: validated alternate tested surface.
  • Artifact (as recorded): TokenRhythm/NeoHorse-1-9B-GGUF@ddcb4c93 :: NeoHorse-1-9B-Q8_0.gguf (sha256 519869730bda973ec50bb3ac42cd874569e5f3a57c3e0d5d23a2b5040e93310f)
  • Runtime (as recorded): llama.cpp b10999.
  • Hardware (as recorded): NVIDIA GeForce RTX 5070 12 GB.
  • Profile Metadata

Events on this configuration:

Chronology

Compact index; full results are in Event results.

DateEventConfigurationStatus
2026-09-25First fresh-model WELP campaign under the model-agentic snapshot: full Model science of the publisher-default Reasoning On profile; case C measured (Reasoning Off sibling required)llama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070READY_WITH_GUARDRAILS
2026-09-25Full WELP characterization of the Reasoning Off deployment profile: independent setup/calibration, same frozen task set, dual-profile completion of the case-C modelllama.cpp official Q8_0 (Reasoning Off, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070READY_WITH_GUARDRAILS
2026-09-25First fresh-model WELP Agentic event: dual-adapter qualification, three required task classes with two scored (1 PASS / 1 FAIL) and the system class INTEGRATION_BLOCKED by a deterministic frozen-harness defectllama.cpp official Q8_0 (Reasoning On = publisher default, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070READY_WITH_GUARDRAILS

Technical provenance

Publication, acceptance and source-pin provenance for each event, collapsed; the scientific content above never depends on it.

Model evidence tree (GitHub)

Event provenance: 2026-09-25 WELP Fresh-Model Characterization: NeoHorse-1-9B (Reasoning On, publisher default)
  • Campaign: neohorse-1-9b-rtx5070-welp-reasoning-on-2026-09-25
  • Export disposition: WEBSITE_PUBLISHED; publication date 2026-09-26
  • Canonical citation (immutable commit/path pin): https://github.com/WumboLabs/evaluations/blob/bf5e29e1afee4f391c4b8f83e77636b887453f53/models/neohorse-1-9b/events/neohorse-1-9b-rtx5070-welp-reasoning-on-20260925/REPORT.md
  • Export identity: event_date 2026-09-25, event_id neohorse-1-9b-rtx5070-welp-reasoning-on-20260925, event_type follow-up, evidence_maturity CURRENT_WELP, model_id neohorse-1-9b, profile_id neohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-on
  • Export source pin: repository-cache WumboLabs/evaluations @ f551e9cc0b00afd5eb8a3bde81ed6f52a12e4e0b path models/neohorse-1-9b/events/neohorse-1-9b-rtx5070-welp-reasoning-on-20260925/website-publication.json sha256 bf4c0f6ad8cb6cef9ea8305576be59f339a466316e15a63a3a18dc61adb6b990

Publication provenance note (as recorded): Native publication of a first fresh-model WELP campaign group under snapshot welp-next-snapshot-2026-09-25-model-agentic; campaign REPORT.md is the primary scientific report; working/raw science retained in local research/model-evaluations/.

Event provenance: 2026-09-25 WELP Reasoning-Profile Characterization: NeoHorse-1-9B (Reasoning Off, supported alternate)
  • Campaign: neohorse-1-9b-rtx5070-welp-reasoning-off-2026-09-25
  • Export disposition: WEBSITE_PUBLISHED; publication date 2026-09-26
  • Canonical citation (immutable commit/path pin): https://github.com/WumboLabs/evaluations/blob/bf5e29e1afee4f391c4b8f83e77636b887453f53/models/neohorse-1-9b/events/neohorse-1-9b-rtx5070-welp-reasoning-off-20260925/REPORT.md
  • Export identity: event_date 2026-09-25, event_id neohorse-1-9b-rtx5070-welp-reasoning-off-20260925, event_type follow-up, evidence_maturity CURRENT_WELP, model_id neohorse-1-9b, profile_id neohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-off
  • Export source pin: repository-cache WumboLabs/evaluations @ 9ecf4f8bd1eed70df5047c64cbe6b1a1b779d1bc path models/neohorse-1-9b/events/neohorse-1-9b-rtx5070-welp-reasoning-off-20260925/website-publication.json sha256 d165d684caa696d516a2b506a0e9c9a8b26ae8ca0a8348918754f6244b5372a3

Publication provenance note (as recorded): Native publication of a first fresh-model WELP campaign group under snapshot welp-next-snapshot-2026-09-25-model-agentic; campaign REPORT.md is the primary scientific report; working/raw science retained in local research/model-evaluations/.

Event provenance: 2026-09-25 WELP Agentic: NeoHorse-1-9B (publisher-default profile, native tool calls)
  • Campaign: neohorse-1-9b-rtx5070-welp-agentic-2026-09-25
  • Export disposition: WEBSITE_PUBLISHED; publication date 2026-09-26
  • Canonical citation (immutable commit/path pin): https://github.com/WumboLabs/evaluations/blob/bf5e29e1afee4f391c4b8f83e77636b887453f53/models/neohorse-1-9b/events/neohorse-1-9b-rtx5070-welp-agentic-20260925/REPORT.md
  • Export identity: event_date 2026-09-25, event_id neohorse-1-9b-rtx5070-welp-agentic-20260925, event_type follow-up, evidence_maturity CURRENT_WELP, model_id neohorse-1-9b, profile_id neohorse-1-9b-q8-0-llamacpp-b10999-rtx5070-deployment-reasoning-on
  • Export source pin: repository-cache WumboLabs/evaluations @ f551e9cc0b00afd5eb8a3bde81ed6f52a12e4e0b path models/neohorse-1-9b/events/neohorse-1-9b-rtx5070-welp-agentic-20260925/website-publication.json sha256 624e4a0197a587ea3c1766901e630d8bb0da337407addc5e798fdf99e0b8e796

Publication provenance note (as recorded): Native publication of a first fresh-model WELP campaign group under snapshot welp-next-snapshot-2026-09-25-model-agentic; campaign REPORT.md is the primary scientific report; working/raw science retained in local research/model-evaluations/.