Applications

Claude Code ollama launch claude --model mannix/gemma4-98e-v7-coderx

Codex App ollama launch codex-app --model mannix/gemma4-98e-v7-coderx

OpenClaw ollama launch openclaw --model mannix/gemma4-98e-v7-coderx

Hermes Agent ollama launch hermes --model mannix/gemma4-98e-v7-coderx

Codex ollama launch codex --model mannix/gemma4-98e-v7-coderx

OpenCode ollama launch opencode --model mannix/gemma4-98e-v7-coderx

Gemma 4 26B-A4B 98e v7-coderx — code-maximal prune

20.8B params · 98 experts (30 dropped) · ~4B active · code-maximal drop map

A research checkpoint that takes Gemma-4-26B-A4B-it and drops ³⁰⁄₁₂₈ experts per layer using a code-maximal recipe on the rebuilt v7 competence maps (audited producers, 10 classes) — generic-code 3× + LiveCodeBench-medium 2× on a [24,40] per-layer floor, with no science or multilingual targeting. Same router, attention, and norms as base, plus the mandatory shared-FFN α=1.2 upweight every coder variant carries.

The strongest coder in the cohort: it spends its whole prune budget on code and lands LiveCodeBench-medium-55 at 98.18% and LCB-100 at 99.0% — the highest of any Gemma-4 prune to date, +1.8pp / +2.0pp past the unpruned 128e (96.36 / 97.0). The trade is graduate science (GPQA 48.48). If you need the science back without giving up the code profile, use the sibling v7-coder (GPQA 70.71, LCB-55 96.36).

Full model card & methodology: ManniX-ITA/gemma-4-A4B-98e-v7-coderx-it on Hugging Face.

Other formats: - GGUF (29 tiers, imatrix, CD-* per-layer mixes + F16 + mmproj): ManniX-ITA/gemma-4-A4B-98e-v7-coderx-it-GGUF - NVFP4A16 (native vLLM, ~13 GB): ManniX-ITA/gemma-4-A4B-98e-v7-coderx-NVFP4A16

Scores (Q6_K, llama.cpp, greedy, same host)

LCB-55	LCB-100	MultiPL-E	HE	HE+	IFEval	GSM8K	MATH-500	AIME	ARC	GPQA-D
98.18	99.00	90.00	95.73	92.68	95.00	91.00	89.00	70.00	94.28	48.48

Reference columns on the same Q6_K run: unpruned 128e LCB-55 96.36 / LCB-100 97.00 / MultiPL-E 90.00; v6-coder LCB-55 92.73 / LCB-100 94.00. v7-coderx tops the cohort on every code/instruction axis; the budget is paid almost entirely on graduate science (GPQA 48.48, vs 128e 67.17).

GGUF tiers (size guide)

Per-tier HE+ was not swept for the v7 cohort — the Q6_K table above is the reference; quality below ~Q3 / 3-bit degrades on the Gemma 4 MoE, so prefer Q4_K_M or higher for production. Full 29-tier list (incl. CD-* per-layer mixes) is on the GGUF repo.

Tier	Size (GB)	bpw	Role
Q8_0	21.16	8.14	near-lossless
Q6_K	17.81	6.84	max fidelity (bench tier)
Q5_K_M	15.07	5.79	high quality
Q4_K_M	13.24	5.09	:latest — recommended default
Q4_K_S	12.21	4.69	compact 4-bit
⭐ IQ4_XS	11.01	4.23	safe sub-12 GB 4-bit
Q3_K_M	10.51	4.04	smallest comfortable
IQ3_XS	9.22	3.54	sub-10 GB
IQ2_M	8.22	3.16	sub-8.5 GB (degraded)
IQ2_S	7.83	3.01	smallest viable

Pull

ollama pull mannix/gemma4-98e-v7-coderx            # :latest = Q4_K_M
ollama pull mannix/gemma4-98e-v7-coderx:IQ4_XS     # safe 4-bit, sub-12 GB
ollama pull mannix/gemma4-98e-v7-coderx:Q6_K       # max fidelity (bench tier)
ollama pull mannix/gemma4-98e-v7-coderx:IQ2_S      # sub-8 GB
ollama pull mannix/gemma4-98e-v7-coderx:vision-Q4_K_M   # + SigLIP vision tower

Inherits Gemma 4’s thinking format — serve with the reasoning parser enabled (--reasoning-format deepseek --reasoning-budget 8192 on llama-server).

Derivative of Gemma 4 — Gemma Terms of Use.

Gemma-4 98e coder max variant, top notch coding skills at the expense of science knowledge

Applications

Models

Readme

Gemma 4 26B-A4B 98e v7-coderx — code-maximal prune

Scores (Q6_K, llama.cpp, greedy, same host)

GGUF tiers (size guide)

Pull