Today's Local LLM Pick: ministral-3:14b on RTX 3090 (2026)

Fast verdict

ministral-3:14b runs at 15.4 tok/s on a 24GB RTX 3090 — in the deliberate range. This model prioritizes quality or parameter count over raw speed. Test it on offline or background tasks first, and consider a smaller quantization if interactive response time matters.

ministral-3:14b fits comfortably in 24GB at standard quantizations. Monitor VRAM usage if you push context beyond 8K tokens. It ranks #9 of 18 in throughput among currently measured models on this RTX 3090. The next faster model is qwq:32b (26.1 tok/s, 70% faster). The next slower model is glm-4.7-flash:bf16 (11.2 tok/s, 37% slower).

The daily goal is simple: help a 3090 owner decide what to download tonight, what to skip, and when a cloud fallback is the better use of time.

Today’s pick

Model: ministral-3:14b
Category: general-purpose
Size tier: medium
Performance tier: deliberate
RTX 3090 speed: 15.4 tok/s
Latency: 8563 ms
Test time: 2026-09-09T07:21:06Z
Baseline command:

ollama run ministral-3:14b

Who should try it

RTX 3090 owners deciding whether to download ministral-3:14b tonight for local experimentation.
Users comparing local inference speed against cloud rental (RunPod, Vast) before committing to a workflow.
Anyone building a local LLM toolbox who wants a verified baseline for this model.

Who should skip it

Users who need long-context production stability before a sustained run has been verified.
Teams whose workload requires predictable p95 latency under concurrency.
8GB/12GB GPU owners unless a smaller quantized variant exists.

Watch points

Workload-specific testing: generic benchmarks do not guarantee performance on your particular use case.
Context length: always test at your target context length before assuming production readiness.
Quantization trade-off: lower quantization saves VRAM but may reduce output quality on nuanced tasks.

Verified benchmark anchors

gpt-oss:20b: 156.1 tok/s | latency 1524 ms | test 2026-04-29T05:39:58Z
qwen3-coder:30b: 84.4 tok/s | latency 7842 ms | test 2026-09-09T07:21:06Z
qwen2.5:14b: 84.0 tok/s | latency 946 ms | test 2026-04-29T05:39:58Z
nemotron-3-nano:30b: 57.0 tok/s | latency 2468 ms | test 2026-04-01T11:53:50Z
deepseek-r1:14b: 52.9 tok/s | latency 2724 ms | test 2026-09-09T07:21:06Z

RTX 3090 decision guide

Offline first: prioritize ministral-3:14b for scheduled batch inference, research, or validation workflows.
Context is the bottleneck: reduce context to the minimum viable length for your task.
Quantize before you buy hardware: Q4 or Q5 may make this viable on 24GB where Q8 is not.
Cloud for interactive: if real-time response is required, treat ministral-3:14b as a cloud-fallback candidate.

Comparisons to validate

ministral-3:14b vs the next-fastest and next-slowest model in the benchmark feed.
ministral-3:14b vs qwen2.5:14b — same size tier, 15 vs 84 tok/s.
ministral-3:14b local power cost vs A100 rental for the same workload.

Next actions

Estimate VRAM fit: /en/tools/vram-calculator/
Model page: /en/models/ministral-3-14b-q4/
Benchmark changelog: /en/benchmarks/changelog/
Local hardware path: /en/affiliate/hardware-upgrade/
Cloud fallback: /go/runpod and /go/vast

Affiliate Disclosure: This post may include affiliate links. LocalVRAM may earn a commission at no extra cost.