Fitting WhisperX large-v3 + a 24B LLM on one 3090: a reproducible context-capping recipe

A developer successfully ran both WhisperX large-v3 (7.7GB) and a 24B parameter LLM (Devstral Small 2) simultaneously on a single 24GB RTX 3090 by reducing the LLM's context window from 40,960 to 8,192 tokens. This cut the KV cache from 6.1GB to 1.25GB, bringing total VRAM usage to 21.9GB and eliminating CUDA out-of-memory errors that occurred when both services overlapped. The fix uses the model's native 8K context length, which covers all real-world triage prompts without quality degradation from rope extrapolation.

This is the technical, reproducible version of a fix I shipped on my own homelab. If you want the narrative version, that's on Medium. This one is the recipe: the measurements, the math, the Modelfile, and the exact prompt I gave Claude Code to generate it. Copy-paste friendly. Repo for the dashboard used throughout: https://github.com/SikamikanikoBG/homelab-monitor https://github.com/SikamikanikoBG/homelab-monitor 18.3 + 7.7 = 26GB → CUDA OOM whenever they overlapped. num ctx to 8192 → KV cache drops from ~6.1GB to ~1.25GB → model footprint ~18.3GB → 14.2 + 7.7 = 21.9GB → both resident, zero OOM, no quality loss. Host: openSUSE, Xeon 56 threads , 125GB RAM, 1x RTX 3090 24GB GPU svc: WhisperX large-v3 speech-to-text GPU svc: Ollama - devstral-small-2 24B, Q4 K M for background email triage Both services run all the time. The OOM only happened when I dictated to my assistant WhisperX while the triage loop was active. nvidia-smi shows instantaneous VRAM. It can't show you which service spiked or when two of them overlapped — and an intermittent OOM is a timing problem. You need per-service VRAM history. I use my own dashboard homelab-monitor for this. The relevant view is "AI Models", which attributes VRAM per model server and per loaded model, over a time range, with OOM markers and a capacity ceiling line. What the history showed at the overlap window: | Service | Peak VRAM | |---|---| | Devstral 24B triage | ~18.3 GB | | WhisperX large-v3 | 7.7 GB | Total | ~26 GB on a 24 GB card | If you want to reproduce the measurement, the dashboard runs as a single container: git clone https://github.com/SikamikanikoBG/homelab-monitor cd homelab-monitor docker compose up -d --build open http://<host :9800 - AI Models / GPU views NVIDIA Container Toolkit required for GPU metrics. Remote hosts are monitored over SSH, no agent. Weights are a fixed cost ~15GB for Devstral 24B at Q4 K M . The variable cost is the KV cache , which scales linearly with num ctx . So the question is: how much context does background email triage actually use? I pulled the request traces from Langfuse. The triage pipeline: Real prompts never exceeded ~5–8k tokens. The model was loaded with a 40k window — ~32k tokens of reserved KV cache doing nothing. Devstral Small is mistral3 . Pull the architecture straight from Ollama: curl -s http://localhost:11434/api/show -d '{"name":"devstral-small-2:latest"}' \ | python -c "import sys,json;mi=json.load sys.stdin 'model info' ;\ print {k:v for k,v in mi.items if 'head count' in k or 'block count' in k or 'length' in k} " Relevant values: block count layers = 40 attention.head count kv = 8 attention.key length = 128 attention.value length = 128 context length native = 8192 rope-extended to 393216 KV cache per token f16 = 2 K+V × layers × kv heads × head dim × 2 bytes : 2 × 40 × 8 × 128 × 2 = 163,840 bytes ≈ 0.156 MB / token So: | num ctx | KV cache f16 | |---|---| | 40,960 | ~6.1 GB | | 16,384 | ~2.5 GB | 8,192 | ~1.25 GB | | 4,096 | ~0.6 GB | 8192 is the sweet spot: it's above the real worst-case prompt ~5–8k and it's the model's native context length, so there's no rope extrapolation quality hit. I rejected 4096 — a 10-email batch with 2k generation can brush up against it. Ollama lets you inherit existing weights and override parameters in a Modelfile, so this costs no extra disk and no re-download. Modelfile.triage : FROM devstral-small-2:latest Native 8K window: covers every triage prompt 10-email batches + 2K generation while keeping the KV cache ~1.25GB so the model + WhisperX fit on one 24GB GPU. PARAMETER num ctx 8192 PARAMETER temperature 0 PARAMETER num predict 2048 SYSTEM """You are a background email-triage engine. Follow the exact output format in each request. Output only the requested label s or field s . Never add explanations, preamble, or commentary. When uncertain, pick the closest valid option. Be terse and deterministic.""" Build it: ollama create devstral-small-2:triage -f Modelfile.triage The optional SYSTEM block is a small bonus: triage prompts want terse, structured output, and pinning that behaviour cuts stray preamble fewer reparse/retry calls = less GPU time . I let Claude Code do the measuring and the Modelfile generation. The prompt, roughly: Analyze my background email triage. Pull the Langfuse traces to find the real prompt/context sizes the triage job uses, decide a safe num ctx cap that won't truncate worst-case batches, confirm the KV-cache savings against the model's actual architecture, and generate an Ollama Modelfile for a context-capped :triage variant. Then tell me the expected VRAM footprint. It came back with: traces show ≤8k tokens, cap at 8192 native window , ~5GB KV saved, expected footprint ~14–16GB. Which matched what the dashboard measured after I deployed it. load it curl -s http://localhost:11434/api/generate \ -d '{"model":"devstral-small-2:triage","prompt":"ping","stream":false}' /dev/null check resident VRAM + context curl -s http://localhost:11434/api/ps \ | python -c "import sys,json; print m 'name' ,round m 'size vram' /1e9,1 ,'GB ctx',m 'context length' for m in json.load sys.stdin 'models' " Result: the triage model holds ~14GB resident at ctx=8192 , down from ~18GB. | Before | After | | |---|---|---| | Triage LLM | ~18.3 GB | ~14.2 GB | | WhisperX large-v3 | 7.7 GB | 7.7 GB | Combined | ~26 GB → OOM | ~21.9 GB → fits | Both services now sit on the card together. Full STT quality, email triage in parallel, ~2GB headroom. No quant change, no CPU offload, no smaller Whisper. nvidia-smi can't diagnose it — get per-service VRAM num ctx to the real workload.Dashboard used for the per-service VRAM history: https://github.com/SikamikanikoBG/homelab-monitor — it's open source, runs in one container, and exists because I needed exactly this view and nvidia-smi wouldn't give it to me.