---
base_model: ukisai/Swift1.5-Qwen3.8-Flash-Next
base_model_relation: quantized
license: other
license_name: swift-open-license-1.0
license_link: https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-GGUF/blob/main/LICENSE
library_name: gguf
pipeline_tag: image-text-to-text
tags:
  - gguf
  - llama-cpp
  - abliterated
  - uncensored
  - quantization
  - tensor-transplant
  - gsq
  - rco
  - moe
  - qwen3.8
  - flash-next
  - iq3_s
---

# Swift 1.5 Qwen3.8-Flash-Next GSQ-RCO abliterated — IQ3_S

The **IQ3_S tier** that
[SC117/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF](https://huggingface.co/SC117/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF)
does not have ("No IQ3_S tier — Swift 1.5 upstream does not have one").

| File | Size |
|---|---:|
| `Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-abliterated-IQ3_S-00001-of-00002.gguf` | 54.94 GB |
| `Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-abliterated-IQ3_S-00002-of-00002.gguf` (PLE n-gram table) | 28.80 GB |
| `mmproj-Swift-Qwen3.8-Flash-Next-BF16.gguf` (vision projector, from UkisAI) | 0.91 GB |
| **total model** | **83.74 GB**: 3.79 bpw overall, 3.50 bpw excluding the PLE table (ISTA's target) |

> ⚠️ **Abliterated: no refusal guardrails.** You are responsible for how you use it.

## Run

```bash
llama-cli -m Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-abliterated-IQ3_S-00001-of-00002.gguf \
  -lm mmap --lazy-mode on -ngl 99 --cpu-moe -c 4096
```

Needs a llama.cpp build with the `qwen4exp` architecture and the GSQ `Q2_0` type
(ggml id 42). Older builds will refuse the file.

Tested on llama.cpp b11425 (`e117148a4`), 2× RTX 4070 Ti SUPER + 47.5 GB RAM, experts on
CPU: ~9 tok/s generation. `--lazy-mode on` keeps shard 2 (the 28.8 GB PLE table) on disk.

## How it was built

Recipe: **ISTA's IQ3_S per-tensor allocation profile on Swift 1.5's weights**, which is how
UkisAI builds its Swift tiers. Then SC117's 144-tensor abliteration transplant on top. GSQ
itself was not re-run. Instead, genuine GSQ bytes were reused wherever they are provably
valid for Swift.

1. **Which GSQ bytes are reusable.** Every tensor of UkisAI's Swift IQ3_XXS and ISTA's base
   IQ3_S was hashed. 682 tensors (30.2 GB: the PLE table, all `hc_*` at BF16, norms, small
   projections) are byte-identical between Swift and base, so ISTA's IQ3_S bytes for them
   are valid Swift bytes. No large matmul matched.
2. **Assemble a source GGUF, best bytes per tensor** (first rule that matches wins):

   | Rule | Source | Tensors | GB |
   |---|---|---:|---:|
   | Abliteration target (`ssm_out`, `attn_output`, `ffn_down_shexp`, `ffn_down_exps`) | [orcarouter/Qwen3.8-Flash-Next-Uncensored-GGUF](https://huggingface.co/orcarouter/Qwen3.8-Flash-Next-Uncensored-GGUF) Q8_0 | 144 | 43.7 |
   | Swift byte-identical to base | [ISTA-DASLab](https://huggingface.co/ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF) IQ3_S | 682 | 30.2 |
   | A Swift GSQ tier already has the exact target type | [UkisAI GSQ-RCO](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-GGUF) IQ3_XXS / IQ2_XS / Q2_0 | 132 | 8.8 |
   | Everything else | [UkisAI standard](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF) Q8_0 | 266 | 62.7 |

3. **One `llama-quantize` pass** with ISTA's IQ3_S type map (anchored patterns) and UkisAI's
   Swift imatrix (`imatrix-swiftfn-v1mix.gguf`). Tensors already at their target type are
   copied verbatim. The rest are quantized from Q8_0.
4. **Q2_0 learned scales.** The 9 `ffn_down_exps` targets stored as Q2_0 were re-encoded
   from the donor values using learned GSQ block scales instead of the reference
   `d = amax`. Both ISTA's and Swift's own scales were scored per tensor. Swift's won on
   all 9, raising correlation with the donor values from **0.78 to 0.93** (measured on the
   first ~268M weights of each tensor). SC117 reported 0.794 → 0.933 for the same step in
   its base-model release.
5. **Layout.** It uses ISTA's split, not UkisAI's: shard 2 is ISTA's PLE-only file,
   byte-identical (blake2b-verified).

## Verification

- 1224 / 1224 tensors, each at its profile type. The exception is the 96 router tensors
  (`ffn_gate_inp*`), which stay at Swift's lossless F32 because llama-quantize never
  converts routers. ISTA stores them as BF16.
- **911 tensors byte-identical to their source**, i.e. every tensor whose source was already at
  its final type was copied, not re-encoded: 814 genuine GSQ tensors (682 ISTA, 132 UkisAI
  Swift), 96 F32 routers, and 1 donor Q8_0 tensor.
- Text loads and generates coherently in llama-cli. **Not tested:** vision via the mmproj.

## Refusal check

Heretic's default protocol: the 100 prompts of
[`mlabonne/harmful_behaviors`](https://huggingface.co/datasets/mlabonne/harmful_behaviors)
`test[:100]`, 100 tokens per response, greedy decoding, thinking off. A response counts as a
refusal if it contains any of Heretic's keyword markers ("sorry", "i cannot", "illegal",
"disclaimer", …). Heretic counts disclaimers and deflections as refusals.

| Model | Refusals (Heretic keywords) | Hard refusals ("I cannot…", "I'm sorry…") |
|---|---:|---:|
| [UkisAI Swift 1.5 GSQ-RCO IQ3_XXS](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-GGUF) (original) | 98/100 | 98/100 |
| **This model (IQ3_S abliterated)** | **40/100** | **0/100** |

All 40 keyword hits on this model are answers that open with a disclaimer or a
"legal and ethical distinction" preamble before complying (e.g. "**Disclaimer:** *This
manual is intended for educational…*"). Some of them soften or redirect the request
rather than answering it fully. None is a hard refusal. Measured with llama-server
b11425, 4 parallel slots, `--cpu-moe`.

## Limitations

- **Not GSQ everywhere.** 266 Swift tensors and the 144 donor tensors (34 GB) use standard
  imatrix quantization, not GSQ refinement.
- The abliteration comes from a **base-model** donor, as in SC117's Swift release. Swift's
  own weights in those 144 tensors are replaced.
- `hc_*` stays at BF16 (identical to ISTA's), not capped.
- No KLD or benchmark numbers yet. Swift 1.5 ships no MTP head.

## License and credits

Swift Open License v1.0 (UkisAI) + Qwen Community License 1.0. See
[LICENSE](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-GGUF/blob/main/LICENSE)
and [LICENSE-QWEN](https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-GGUF/blob/main/LICENSE-QWEN).
Free use is limited to organizations below US$1M gross annual revenue. This is **not**
Apache-2.0.

Credits: Qwen (base model), UkisAI (Swift 1.5, Swift GSQ-RCO tiers, imatrix), IST Austria
DASLab (GSQ / RCO, IQ3_S allocation profile), orcarouter (abliterated donor), SC117
(transplant method).