WUMBOLABS / EVALUATIONS CANONICAL MODEL EVALUATION

/evaluations/granite-4.2-8b

Granite 4.2 8B

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

latest evidence 2026-09-24

Overview

IBM (ibm-granite) · 1 tested configuration · 1 evidence event · newest evidence 2026-09-24.

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: NOT_READY - attributed to the 2026-09-24 event, configuration llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070.
  • Recommended configuration: llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K) (current): event result.
  • Latest evidence: 2026-09-24: event result
  • 2026-09-24 · llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070: classification NOT_READY. 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.

2026-09-24: First published evaluation via linked completion of the methodology-repair predecessor (official Q4_K_M, upstream llama.cpp b10999)

WELP Initial Evaluation: IBM Granite 4.2 8B (reasoning-on) · config llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070 · maturity: CURRENT_WELP · status: NOT_READY

Methodology snapshot: welp-next-snapshot-2026-09-24-context-outcome-repair · WELP commit ead28bde1ea9bce20cec6b3930125016068a8bbc

Configuration: llama.cpp official Q4_K_M (reasoning-on, 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)

SUBMITTED/NEW, service record cmug9qe0i0d5wlq01lep45uua (APPROVED, verifiedRun false)

Tested condition: llama.cpp b10999 · RTX 5070.

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

Identity

FieldValue
Profile identitygranite-4.2-8b-q4-k-m-llamacpp-b10999-rtx5070-deployment
ModelGranite 4.2 8B
Producernot recorded
Tested artifactnot recorded
Precisionnot recorded
Artifact SHA-25616a9369d0805f80b7377d25d87f937a90c05dc04ad79173a52001e42c9aab311
Campaigngranite-4.2-8b-rtx5070-welp-context-outcome-completion-2026-09-24
Record date2026-09-24

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 Q4_K_M (reasoning-on, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070Q4_K_M official publisher GGUFllama.cpp b10999RTX 5070not recorded1 (2026-09-24)
Configuration details: llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070
  • Full profile name: llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K).
  • Profile identity: granite-4.2-8b-q4-k-m-llamacpp-b10999-rtx5070-deployment.
  • Status: current canonical/recommended tested surface.
  • Deployment topology (as recorded): Full-GPU upstream llama.cpp CUDA serving on RTX 5070 12GB: -ngl 99 (40/40 layers), -fa on, f16 KV, -np 1, embedded GGUF Jinja template, DEPLOYMENT sampler (temp 0.2 / top_k 80 / top_p 1.0 / repeat_penalty 1.05); effective reasoning ON (built-in thinking; OFF control proven effective but not the campaign profile). Default context 32,768 / native exact 131,072; uncached scientific arms via --cache-ram 0 --no-cache-prompt --slot-prompt-similarity 0 with cache_prompt:false (proven by repeated-prompt probe). Publisher sampler (temp 1.0 / top_p 0.95) measured as a matched comparison, never pooled.
  • Artifact (as recorded): ibm-granite/granite-4.2-8b-GGUF@93f3f6a8 :: granite-4.2-8b-Q4_K_M.gguf (sha256 16a9369d0805f80b7377d25d87f937a90c05dc04ad79173a52001e42c9aab311)
  • 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
2026-09-24First published evaluation via linked completion of the methodology-repair predecessor (official Q4_K_M, upstream llama.cpp b10999)llama.cpp official Q4_K_M (reasoning-on, deployment sampler, 32K) · llama.cpp b10999 · RTX 5070NOT_READY

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: 2026-09-24 WELP Initial Evaluation: IBM Granite 4.2 8B (reasoning-on)
  • Campaign: granite-4.2-8b-rtx5070-welp-context-outcome-completion-2026-09-24
  • Export disposition: WEBSITE_PUBLISHED; publication date 2026-09-24
  • Canonical citation (immutable commit/path pin): https://github.com/WumboLabs/evaluations/blob/adcfa178b6dd9b52eede2c433ffc691dbcd26941/models/granite-4.2-8b/events/granite-42-8b-rtx5070-welp-context-completion-20260924/REPORT.md
  • Export identity: event_date 2026-09-24, event_id granite-42-8b-rtx5070-welp-context-completion-20260924, event_type initial-evaluation, model_id granite-4.2-8b, predecessor_campaign granite-4.2-8b-rtx5070-welp-characterization-2026-09-24, profile_id granite-4.2-8b-q4-k-m-llamacpp-b10999-rtx5070-deployment, profile_status current
  • Export source pin: repository-cache WumboLabs/evaluations @ 250afdf042677d8c6dc19275c357ad0f5a2b870a path models/granite-4.2-8b/events/granite-42-8b-rtx5070-welp-context-completion-20260924/website-publication.json sha256 978d09e20e166ff7ecec559a1b10a5a7d3de10292aa663ac3e6ce6e018cced1b

Publication provenance note (as recorded): Native current-WELP linked completion event under the consolidated architecture. The retained COMPLETE_WITH_GAPS predecessor campaign is preserved unchanged and unpublished; every raw evidence file is reused byte-identically (size + SHA-256 verified); only the answerless context cell's disposition/outcome is re-derived from original raw bytes under welp-context 0.3.0-draft.