qwen3.5-9b-gguf
Consumer-supplied Wekala model-library assets derived from the attributed official publisher. Models and this repository are distributed separately from the Wekala SDK.
Origin and licence
| Field | Value |
|---|---|
| Original developer | Qwen Team, Alibaba Cloud |
| Original model | Qwen/Qwen3.5-9B |
| Pinned source revision | c202236235762e1c871ad0ccb60c8ee5ba337b9a |
| Original publication | 2026-03-02 (day) |
| Date evidence | Publisher source |
| Licence | apache-2.0; preserved in LICENSE |
| Full source parameters | 9,653,104,368 (includes embeddings and source modalities) |
| Nile Academy role | Official artifact mirror or explicitly recorded format conversion; no additional training |
Original publisher cards and available notices are preserved in provenance/. The original model licence governs the weights; this repository grants no rights to the Wekala SDK.
Files and quantization
Q8_0 is the higher-precision comparison; Q6_K/Q5_K_M trade memory for precision; Q4_K_M needs paired task evaluation. No general quality-loss or 8GB-fit qualification yet.
| File | Quantization | Size (GB) | Origin |
|---|---|---|---|
| Qwen3.5-9B-Q4_K_M-llamacpp-4413a19.gguf | Q4_K_M | 5.780 | existing Nile conversion; original lineage retained |
Exact file SHA-256 values, upstream paths/revisions and conversion receipts are recorded in model-metadata.json.
Published benchmarks
Publisher-reported source-model scores follow. They are not Wekala measurements of these quantized files. Harness, precision, reasoning settings and task versions differ; do not treat them as a normalized cross-model ranking.
| Benchmark | Score (%) | Mode / settings |
|---|---|---|
| MMLU-Pro | 82.5 | publisher-default; Publisher card protocol; cross-model scores are not normalized. |
| IFEval | 91.5 | publisher-default; Publisher card protocol; cross-model scores are not normalized. |
| BFCL-V4 | 66.1 | publisher-default; Publisher card protocol; cross-model scores are not normalized. |
Qualification and use
Existing Windows development routes have separate readiness observations. Broad quality, sustained device fit, mobile/browser and owner acceptance remain separate.
Wekala broad quantization-quality qualification: not qualified. No general “no significant quality loss” claim is made. Official QAT or publisher comparisons are attributed separately. Before recommendation, use paired reference/variant evaluation across instruction, knowledge, math, multilingual/Arabic and tool tasks under the registered Wekala campaign.
Device fit is separate: weight size excludes context state, scratch buffers and driver headroom. CPU and CUDA are execution backends for the same weight file, not separate models. Mobile/browser and sustained 8GB-VRAM operation require their own measured profiles.
Models can produce incorrect or unsafe outputs. Tool calls are proposals subject to the consuming application’s permissions and controller admission.
Original-precision checkpoint
The original BF16 checkpoint is preserved under original/c202236235762e1c871ad0ccb60c8ee5ba337b9a/, copied from Qwen/Qwen3.5-9B at the exact pinned revision. Original tensor files total 19,306,310,880 bytes. This includes the source modalities; publisher executable Python is excluded. Source licences, original release date and published benchmarks above still apply.
| Download | Precision | Bytes | Scope |
|---|---|---|---|
| model-original-bf16.gguf | BF16 original precision | 18,407,321,440 | Text-only GGUF reference; separate from the complete original checkpoint |
Original-precision options have no 8 GB eligibility ceiling. File size is not total RAM/VRAM usage. These are not Q8 and are not dequantized QAT/Q4 weights. GGUF conversion fidelity, runtime profiles and sustained device fit remain separate qualification gates; no lossless-conversion or quality guarantee is implied. See the immutable source and reference record.