Skip to content

Magistral Small 1.2 24B

24B

Mistral Magistral

Mistral's reasoning model on a Mistral Small 3.2 base, with reasoning wrapped in [THINK] tags. At Q4_K_M it needs about 15.7 GB at 4K, tight on a 16 GB card and comfortable on 20 GB — explicit chain-of-thought a level below the 70B-class reasoners most guides assume.

⬇ 1.5K HF downloads♥ 10 likesbartowski/mistralai_Magistral-Small-2509-GGUF· stats from 11/22/2026
Consumer GPUMac / Apple Silicon

131K

Max Context

5

Quant Variants

GGUF Q6_K

Best Quality

+0.6%

Perplexity rise (best build)

Quantization Variants

Per-quant VRAM, quality loss, and inference speed on RTX 4090

Measured = run on this site · Estimated = not run here: calculated, or a figure this site has not verified · Community = public reports

FormatLevelBPWVRAMPPL LossSpeedSourceActions
GGUFQ4_K_M4.8615.7 GB2.9%60 tok/sCommunity
CalcHF
GGUFQ5_K_M5.6918.2 GB1.4%54 tok/sEstimated
CalcHF
GGUFQ6_K6.5620.9 GB0.6%47 tok/sEstimated
CalcHF
AWQINT4413.0 GB4.0%76 tok/sEstimated
CalcHF
EXL24.65bpw4.6515.0 GB2.5%—Estimated
CalcHF

Which file each size is for

A file size is not a memory figure: the VRAM column is the calculator's total at 4K context — weights plus KV cache plus a 10% activation buffer. Units differ on purpose: file sizes are decimal GB, as Hugging Face lists them; the VRAM column is GiB (1024³ bytes) with 2% added to the weights, so 7.12 GB of file counts as about 6.8 GiB there. A vision projector (mmproj), where the repo ships one, is a separate download and is in neither number.

Related community runs

These runs describe their own settings. Their speeds do not replace this estimate or prove your configuration fits.

Our reviewed link index has no run for this model yet. You can browse community reports or share your own test below.

Browse more runs on llamaPerf ↗
30B-A3B

Qwen3-VL 30B-A3B Instruct

Alibaba Qwen3-VL

Consumer GPUPro GPU
Q4_K_M19.9 GB+2.6%perplexity rise

Multimodal MoE with only ~3B active parameters, so it stays responsive on Apple unified memory and survives CPU offload far better than a dense 30B. At Q4_K_M it needs about 19.9 GB at 4K, comfortable on a 24 GB card; Q3_K_M brings that to 15.1 GB. The vision encoder is a separate 1.1 GB file.

26B-A4B

Gemma 4 26B-A4B IT

Google Gemma 4

Consumer GPUPro GPU
Q4_K_M18.2 GB—perplexity rise

Google's Gemma 4 mixture-of-experts model: about 25B parameters in the text model, roughly 4B active per token, a 256K window and image input. Only 5 of its 30 layers keep a cache that grows with context, and those use fewer, wider heads, so the cache stays small — about 0.9 GB at 32K and 5.3 GB at the full 256K. At Q4_K_M that is about 18.8 GB at 32K: comfortable on a 24 GB card, which can still load the whole window (tight, about 23.6 GB). A 16 GB card cannot hold it, since the Q4_K_M file alone is 17.0 GB. The vision encoder is a separate 1.2 GB file on top. Sizes come from bartowski's GGUF files, whose Q4_K_M works out at 5.40 bits per weight, well above the generic 4-bit rate; Unsloth's UD-Q4_K_M is a different file, 0.09 GB smaller. No per-level quality loss has been published.

31B

Gemma 4 31B IT

Google Gemma 4

Consumer GPUPro GPU
Q4_K_M22.1 GB—perplexity rise

Google's dense Gemma 4 flagship: about 31B parameters, a 256K window and image input. Only 10 of its 60 layers cache the full context, and those use four wide KV heads, so 128K of context adds about 11 GB of cache where a conventional 60-layer design would need about 120 GB. At Q4_K_M it needs about 22.1 GB at 4K, tight on a 24 GB card; Q3_K_M is about 18.3 GB at 4K. A 32 GB card holds Q4_K_M at 32K comfortably. Google also publishes its own quantization-aware (QAT) Q4_0 build, a single 17.7 GB file: about 20.1 GB at 4K, comfortable on 24 GB up to 16K. Its 1.2 GB vision encoder is a separate file, and with it loaded even 4K is tight on 24 GB. The other levels are sized from bartowski's GGUF files, and no per-level quality loss has been published. That repository also carries two MTP draft-head files (349 MB at Q4_0, 515 MB at Q8_0) for speculative decoding; they are not the model itself and are not counted in these sizes.

30B

Muse Glimmer 30B

Meta Muse

Consumer GPUMac / Apple Silicon
Q4_K_M17.7 GB—perplexity rise

Meta's first open-weight model since Llama 4, released in August under Apache 2.0 and built to run on one card. It is a dense model: a 28B text decoder that reads images through a separate vision encoder and answers in text. Three of every four layers use a 2K sliding window and the rest share just 2 KV heads, so long context is unusually cheap. At Q4_K_M it needs about 17.7 GB at 4K context and only 19.4 GB at its full 128K window, comfortable on a 24 GB card either way. Q5_K_M is tight on 24 GB, and Q8_0 wants a 32 GB card. The vision encoder (about 1.4 GB as a GGUF mmproj file) and Meta's optional speculative-decoding drafter load on top and are not included. Sizes come from the published GGUF file sizes; the Q4_K_M row is Meta's own official build. Meta's benchmark claims are its own, and no per-level quality loss has been published.

Running Magistral Small 1.2 24B locally

At Q4_K_M and 4K of context, Magistral Small 1.2 24B needs about 15.7 GB — 13.6 GB of weights, 0.63 GB of KV cache and a 1.4 GB activation buffer. The smallest card in this index that clears that comfortably is the Radeon RX 7900 XT at 20 GB, and 35 of the 105 cards here do. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.

What longer context costs

Going from 4K to 32K adds about 4.37 GB, taking the total to 20.5 GB. Weights do not move with context — only the KV cache does, and it grows linearly, so this is the number to watch when planning for long documents. The model's native window is 128K; holding all of it at Q4_K_M would need about 37 GB.

Which build to download

This index tracks 3 formats for it — GGUF, AWQ, EXL2 — across 5 levels. Q6_K carries the lowest published perplexity loss at 0.6%. The fastest level listed is AWQ INT4 at about 76 tok/s on an RTX 4090, batch 1 — an estimate, not a run on this site's hardware. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 — CUDA only, and now archived, so local use rather than serving.

Common questions

How much VRAM does Magistral Small 1.2 24B need?
About 15.7 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 20.5 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.
Will Magistral Small 1.2 24B run on a 20GB GPU?
Yes — at Q4_K_M and 4K context it needs about 15.7 GB, which leaves 4.3 GB spare on a Radeon RX 7900 XT. That is the smallest card in this index that clears it comfortably; 35 of 105 do.
Which quantization of Magistral Small 1.2 24B should I use?
Q6_K has the lowest published quality loss (0.6%), and Q4_K_M is the level most people run. All 5 levels in the index are Q4_K_M, Q5_K_M, Q6_K, AWQ INT4, EXL2 4.65bpw.

Where this model fits

Sized at Q4_K_M with a 4K context window, smallest card first. Comfortable means the estimate uses at most 88% of the memory.

Ships in
GGUFAWQEXL2

Cards for this model

This page's figures change when the model or the runtime does.Last updated 2026-11-22 RSS → /feed.xml