Mistral Large Instruct 2407 — Hardware Requirements & GPU Compatibility
ChatFunctions
Mistral-Large-Instruct-2407, also known as Mistral Large 2, is Mistral AI's flagship dense instruction-tuned model of about 123 billion parameters, designed to run efficient single-node inference despite its size. It targets state-of-the-art reasoning, coding, and knowledge tasks with native function calling and JSON output for agentic use, and covers a dozen or more natural languages including French, German, Spanish, Chinese, Japanese, Russian, and Korean alongside more than 80 programming languages. Mistral reports reduced hallucination rates and stronger reasoning compared with the original Mistral Large. Even tuned for single-node deployment, a dense model of this size needs a multi-GPU workstation to run. Context length is 131,072 tokens (a 128k window). It is released under the Mistral AI Research License, a custom license restricting use to non-commercial research purposes; commercial deployment requires a separate license from Mistral AI. It was published in July 2024.
Mistral Large Instruct 2407 runs on a 96 GB+ Mac or a workstation GPU
Best local pick → Mac Studio (M3 Ultra, 96GB) · Q4_K_M · 86.4 GB · ~7 tok/s
Macs from 48 GB (IQ2_XXS · ~12 tok/s)
Score 81.7#87 of 106 open#44 of 62 that fit ≤96 GBOther
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
llmrun Composite Score →
81.7
- 90% range 77.1–86.3
- #87 of 106 open models
- #44 of 62 that fit ≤96 GB
Run Mistral Large Instruct 2407
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/MaziyarPanahi/Mistral-Large-Instruct-2407-GGUF:Q4_K_MQ4_K_M · needs ~86.4 GB · 16K context
Download: Mistral-Large-Instruct-2407.Q4_K_M.gguf-00001-of-00007.gguf73.22 GB + 6 partsMaziyarPanahi
Ways to runOther ways
each opens step-by-stepRun in cloud
Your own private copy · Q4_K_M · 96 GB GPU
- The easiest way to try this model
- Running in a few minutes, easy setup
- Chat right here or connect any app
$5+ bonus for new users
Runs on B200 (192 GB) (105 GB headroom) · Q4_K_M
- Generation speed
- ~64 tok/s
- generation speed
- Cost per 1M output tokens
- $25.83
- per 1M output tokens
How Much VRAM Does Mistral Large Instruct 2407 Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 57.3 GB | 69.1 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Does not fit: needs 62.8 GB | |
| 59.0 GB | 70.8 GB | 3-bit small — 3-bit small quantization | Does not fit: needs 64.5 GB | |
| 65.8 GB | 77.6 GB | 3-bit medium — 3-bit medium quantization | Does not fit: needs 71.3 GB | |
| 67.4 GB | 79.3 GB | 4-bit · legacy — 4-bit legacy quantization | Does not fit: needs 73 GB | |
| 80.9 GB | 92.7 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Does not fit: needs 86.4 GB | |
| 96.1 GB | 107.9 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Does not fit: needs 101.6 GB | |
| 111.3 GB | 123.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 116.8 GB | |
| 134.9 GB | 146.7 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 140.4 GB |
Selected Q4_K_M
4-bit medium quantization86.4GBat 16K context
- Weights
- 79.8 GB
- KV cache
- 6.3 GB
- Overhead
- 0.3 GB
Doesn't fit RTX 4090 (reference) · needs 62.4 GB more
Or .
Which Hardware Can Run Mistral Large Instruct 2407?
Q4_K_M · 80.9 GBMistral Large Instruct 2407 (Q4_K_M) requires 80.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 106+ GB is recommended. No consumer GPU has enough memory. 30 devices with unified memory can run Mistral Large Instruct 2407, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio (M3 Ultra, 96GB).
48 GB
48 GB tier →Q4_K_M · best fit IQ2_S (47.7 GB)
2 Mac classes · 1 device family
M5 Max48 GB~10Simulate output on Mac Studio M5 Max (48 GB)M5 Pro48 GB~5Simulate output on Mac Mini M5 Pro (48 GB)Snapdragon X2 Elite Extreme (X2E-96-100)~4Simulate output on Snapdragon X2 Elite Extreme Copilot+ PC96 GB
fits · 9.6 GB free · up to 32K ctx
2 Mac classes
M3 Ultra*96 GB~7Simulate output on Mac Studio (M3 Ultra, 96GB)M3 Max*96 GB~3Simulate output on MacBook Pro 14" M3 Max (96 GB)
By VRAM tier · typical GPUs and Macs at Q4_K_M
How good is Mistral Large Instruct 2407?
The llmrun Score merges 6 public benchmark boards into a single figure, where 100 represents the average for open models.
Score vs VRAM in the ≤96 GB tier
- Mistral Large Instruct 2407
- Other models
- Best Score for its VRAM
#44 of 62 that fit ≤96 GB
Each dot is a model at the best quant that fits ≤96 GB. The line steps up to the best Score available at each amount of VRAM; a dot below it is beaten by a model that needs no more memory.
Full ≤96 GB ranking →Benchmarks
The 6 boards that weigh most in the llmrun Score, of 9 with a result for this model; badges show who ran the test.
- GPQA Diamond
Epoch AI · Jul 2024Independent
Independent49.0%
#58/100 open
- LMArena Text
LMArena · Oct 2026Arena
Arena1314.4
#99/188 open
- AIME 2024/2025
Epoch AI · Jul 2024Independent
Independent8.5%
#61/88 open
- MMLU-Pro
TIGER-LabIndependent
Independent65.9%
#45/146 open
- SimpleBench
Epoch AI · Jul 2024Independent
Independent22.5%
#21/25 open
- MATH Level 5
Epoch AI · Jul 2024Independent
Independent44.8%
#17/39 open
In the ≤96 GB class
62 models fit ≤96 GBSame Score, only models that fit ≤96 GB at their best quant: the ranking a ≤96 GB owner actually chooses from.
- #1
MiMo V2.6 Flash RLXiaomi
131.9
~153 tok/s · 93.8 GB
- #2
DeepSeek V4 Flash 0731DeepSeek
127.6
~103 tok/s · 95.5 GB
- #3
Qwen3.8 27BAlibaba
122.0
~67 tok/s · 17.9 GB
- #4
Inkling SmallThinking Machines
119.6
~356 tok/s · 92.9 GB
- #5
DeepSeek V4 FlashDeepSeek
115.3
~162 tok/s · 91.4 GB
- #6
MiniMax M2.5MiniMax
111.8
~139 tok/s · 93.9 GB
Alternatives to Mistral Large Instruct 2407
Mistral family · 35 models →Official models with an llmrun Score, each answering one question; numbers are at the best quant that fits a 24 GB GPU.
Best alternative at this size
Mistral Medium 3.5 128B
Mistral AI
- 20.4 higher Score
- 3.6 GB less VRAM
Score 102.1 · 82.8 GB
Other vendor
Inkling Small
Thinking Machines
- 37.9 higher Score
- 76.3 GB more VRAM
Score 119.6 · 162.7 GB
Similar models
10| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| MiniMax M2.5MiniMax | 111.8 | 139.6 GB | —needs > 24 GB | similar score |
| Qwen3.5 122B A10BAlibaba | 109.4 | 75.5 GB | —needs > 24 GB | similar score |
| Qwen3 235B A22B Thinking 2507Alibaba | 108.4 | 142.9 GB | —needs > 24 GB | similar score |
| MiniMax M2.7MiniMax | 107.4 | 139.6 GB | —needs > 24 GB | similar score |
| MiniMax M2.1MiniMax | 106.2 | 139.6 GB | —needs > 24 GB | similar score |
| MiniMax M2MiniMax | 104.0 | 139.6 GB | —needs > 24 GB | similar score |
| Phi 4Microsoft | 83.9 | 12.5 GB | ~69 | similar score |
| Qwen2.5 72B InstructAlibaba | 82.2 | 49.3 GB | —needs > 24 GB | similar score |
| Llama 3.3 70B InstructMeta | 80.0 | 51.3 GB | —needs > 24 GB | similar score |
| Gemma 3 27B ITGoogle | 79.4 | 21.7 GB | ~36 | similar score |
Specifications
- Publisher
- Mistral AI
- Family
- Mistral
- Parameters
- 122.6B
- Release Date
- 2024-07-24
- License
- Other
Frequently Asked Questions
How much VRAM does Mistral Large Instruct 2407 need?
Mistral Large Instruct 2407 requires 80.9 GB of VRAM at Q4_K_M, or 269.7 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 122.6B × 4.8 bits ÷ 8 = 73.6 GB
KV Cache + Overhead ≈ 7.3 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Mistral Large Instruct 2407?
No — Mistral Large Instruct 2407 requires at least 37.1 GB at IQ2_XXS, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
What's the best quantization for Mistral Large Instruct 2407?
For Mistral Large Instruct 2407, Q4_K_M (80.9 GB) offers the best balance of quality and VRAM usage. Q5_0 (84.3 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 37.1 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Mistral Large Instruct 2407 on a Mac?
Yes — Mac Studio M5 Max (48 GB) and 55 other Macs can run Mistral Large Instruct 2407, the smallest at IQ2_XXS. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 80.9 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio (M3 Ultra, 96GB).
Can I run Mistral Large Instruct 2407 locally?
Yes — Mistral Large Instruct 2407 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Mistral Large Instruct 2407?
At Q4_K_M, Mistral Large Instruct 2407 can reach ~59 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 80.9 × 0.65 = ~64 tok/s
Estimated speed at Q4_K_M (80.9 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Mistral Large Instruct 2407?
At Q4_K_M, the download is about 73.57 GB. The full-precision BF16 version is 245.22 GB. The smallest option (IQ2_XXS) is 33.72 GB.
Which GPUs can run Mistral Large Instruct 2407?
No single consumer GPU has enough VRAM to run Mistral Large Instruct 2407 at Q4_K_M (80.9 GB). Multi-GPU or professional hardware is required.
Which devices can run Mistral Large Instruct 2407?
30 devices with unified memory can run Mistral Large Instruct 2407 at Q4_K_M (80.9 GB), including ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB), GMKtec EVO-X2 (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.