MistralRetired
Mixtral 8x7B Instruct
Mistral's Mixtral announcement says 46.7B total parameters and 12.9B active. The later docs page rounds this to 47B and 13B.
The Mixtral announcement and the model docs say 32k context.
From the model card
Strengths
- The announcement describes a sparse mixture of experts: 8 experts, 2 active per token, Apache-2.0.
- It lists English, French, Italian, German, and Spanish, and it presents the model as strong at code.
Weaknesses
- Mistral's docs say this model was retired on 30 March 2025. The listed replacement is Mistral Small.
- Memory follows 46.7B. All experts are stored.
- No Frankenstein row yet.
Sources: Mixtral announcement · Mixtral 8x7B docs (retirement note) · Instruct model card · config.json
Memory at 8,192 tokens
Computed, not measured. Weights use llama.cpp’s whole-file averages (Q4_K_M = 4.8944 bits/weight, Q8_0 = 8.5008) on the total parameter count. KV is fp16 and does not depend on the quant. The formula, including the sliding-window and latent-KV cases, is on the methodology page.
| Q4_K_M | Q8_0 | |
|---|---|---|
| Weights | 26.61 GiB | 46.22 GiB |
| KV at 8,192 | 1 GiB | 1 GiB |
| Total | 27.61 GiB Computed | 47.22 GiB Computed |
32 layers, 8 KV heads, head dim 128. Every layer stores the full context.
Tokens per second
Not yet measured. This lab has no Frankenstein row for Mixtral 8x7B Instruct, so there is no estimated tok/s either.
Hardware tiers
Minimum is the smallest RAM that holds Q4_K_M plus 8k KV. Suggested does not name a second machine or a speed this lab has not run. Performance means a CUDA-class card with 12 GB or more of GDDR6 or GDDR6X and the weights fully on the card. The Tesla P40 is 24 GB of GDDR5, so a P40 speed is not the performance-tier number.
| Tier | Example hardware | Quant | Fit | tok/s |
|---|---|---|---|---|
| Minimum | CPU-only, with at least 29.61 GiB of system RAM for the Q4_K_M weights, 8k KV, and a 2 GiB planning allowance. | Q4_K_M | CPU-only, if the RAM is there | CPU: not yet measured |
| Suggested | A machine you are willing to wait on, or a GPU that can hold the quant. This lab has not measured a separate “comfortable” machine, so this row does not invent one. | Q4_K_M until a Q8 run exists | Whichever of the machines above actually holds the weights | not yet measured |
| Performance | A CUDA-class card with enough GDDR6 or GDDR6X to hold about 28.61 GiB (Q4_K_M weights, 8k KV, and a 1 GiB planning allowance). 12 GB is not enough for that total. | Q4_K_M | Full VRAM only if the card is larger than 12 GB | not yet measured |
Where it sits
Best-for is a judgment from the card’s stated jobs and the model’s size. It is not a speed ranking. The matching build is CPU-only used Xeon. Score it against a machine you have on What can my rig run?