Gemma 4 26B-A4B · vLLM · NVFP4 + EAGLE3 · conc 128
status: done
Configuration
| Model | google/gemma-4-26B-A4B-it |
|---|---|
| Company | |
| Family | Gemma |
| Parameters | 26B / 4B (MoE) |
| Engine | vLLM + EAGLE3 (speculative decoding) |
| Quant / precision | NVFP4 |
| Why this quant | NVIDIA NVFP4 base (modelopt) + RedHatAI's official EAGLE3 speculator (RedHatAI/gemma-4-26B-A4B-it-speculator.eagle3) — the EAGLE3 line of the crossover figure. c128 point of the beyond-c32 ceiling-finding extension. |
| Download | nvidia/Gemma-4-26B-A4B-NVFP4 |
| Context window | 65536 |
| Input modalities | text, image (served text-only here) |
Measured results
| Prefill tok/s | 1384.59 |
|---|---|
| Decode tok/s | 1210.22 |
| Peak memory (GB) | 111.88 (system MemAvailable delta (10s sampling) — vLLM static KV reservation (util 0.85) + EAGLE3 head + draft buffers; ~6.6 GB free at peak) |
| Completed | 2026-07-04 13:20 +0800 |
Full run command
# Run under scripts/mem-watchdog.sh (floor 3.5 GB) — GB10 unified memory; see notes/BENCHMARKING.md.
VLLM_IMAGE=vllm/vllm-openai:nightly-aarch64 scripts/bench-vllm-serving.sh nvidia/Gemma-4-26B-A4B-NVFP4 65536 128 1000 900 256 \
--speculative-config '{"model":"RedHatAI/gemma-4-26B-A4B-it-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}'
# 1000/1000 prompts (clean full run), 0 errors, 194.6s. ready after 287s. TTFT median 562.6 ms, TPOT median 94.0 ms, req thr 5.138/s. Watchdog no trip (6.6 GB free).
# SpecDecoding: mean acceptance length 2.11 · avg draft acceptance 37% · per-position ~0.59/0.34/0.19.
Notes
Decode 1210.22 tok/s aggregate at concurrency 128 — still climbing. NVFP4 + RedHatAI EAGLE3 speculator, second point of the beyond-c32 ceiling-finding sweep (context held at 65536).
- Compute still has headroom: decode 1210.22 tok/s vs c64 (888.98) = 1.36× per doubling — matching base’s deceleration curve. EAGLE3 tracks between base and MTP throughout (base 995 · EAGLE3 1210 · MTP 1380 at c128).
- Memory tightening (like MTP, unlike base): peak 111.88 GB delta = ~6.6 GB free — the EAGLE3 draft head + per-seq draft buffers climb with concurrency (base sat at ~9.4 GB here). This is why c256 is the danger point for the spec lines but not for base. Watchdog (floor 3.5 GB) armed, did not trip at c128.
- Acceptance flat ~37%: mean acceptance length 2.11, per-position 0.59/0.34/0.19 — consistent with c1–c64; the weak-but-stable EAGLE3 draft.
- Image (pinned):
vllm/vllm-openai:nightly-aarch64@sha256:e414712fdc04…. Ready after 287 s. - Workload: ShareGPT V3, concurrency 128. 1000/1000, 0 errors in 194.6 s (no time cap). TTFT median 562.6 ms, TPOT median 94.0 ms, req throughput 5.138/s.
</content> </invoke>