Llama 3.3 70B Instruct vs Mistral Large Instruct 2407
Llama 3.3 70B Instruct and Mistral Large Instruct 2407 match in llmrun Score, with values of 79.7 versus 81.1; neither fits 24 GB.
Model A · Meta
Llama 3.3 70B Instruct70.6B · 128K context · llama3.3 · Nov 2024
Model B · Mistral AI
Mistral Large Instruct 2407122.6B · Other · Jul 2024
Choose Llama 3.3 70B Instruct if you want…
Fits a 48 GB GPU
Q4_K_M · 51.3 GB; Mistral Large Instruct 2407 needs 42.6 GB at its smallest quant
Fits a Mac 64 GB
Q4_K_M · 51.3 GB; Mistral Large Instruct 2407 needs 42.6 GB at its smallest quant
Lighter on memory
46.6 vs 80.9 GB at Q4_K_M
Newer release
Nov 2024 vs Jul 2024
Choose Mistral Large Instruct 2407 if you want…
Ahead on the benchmarks
5 of 7 shared boards
On your hardware
Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.
8 GB
RTX 4060 · 8 GB tier
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
Mistral Large Instruct 2407
Doesn't fit
smallest quant needs 42.6 GB
12 GB
RTX 3060 12GB · 12 GB tier
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
Mistral Large Instruct 2407
Doesn't fit
smallest quant needs 42.6 GB
16 GB
RTX 5060 Ti 16GB · 16 GB tier
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
Mistral Large Instruct 2407
Doesn't fit
smallest quant needs 42.6 GB
24 GBreference
RTX 4090 · 24 GB tier
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
Mistral Large Instruct 2407
Doesn't fit
smallest quant needs 42.6 GB
32 GB
RTX 5090 · 32 GB tier
Llama 3.3 70B Instruct
IQ2_M · ~45 tok/s
30.9 GB · up to 16Kfits
Mistral Large Instruct 2407
Doesn't fit
smallest quant needs 42.6 GB
48 GB
RTX A6000 · 48 GB tier
Llama 3.3 70B Instruct
Q4_K_M · ~11 tok/stight
51.3 GB · up to 4K
Mistral Large Instruct 2407
IQ2_S · ~12 tok/s
47.7 GB · up to 16Kmore headroom
Mac 32 GB
Mac Mini M4 · Macs
Llama 3.3 70B Instruct
IQ2_XS · ~4 tok/s
28 GB · up to 16Kfits
Mistral Large Instruct 2407
Doesn't fit
smallest quant needs 42.6 GB
Mac 64 GB
Mac Studio M4 Max · Macs
Llama 3.3 70B Instruct
Q4_K_M · ~8 tok/s
51.3 GB · up to 32Kmore headroom
Mistral Large Instruct 2407
Q2_K_S · ~7 tok/s
59.5 GB · up to 16K
Mac 128 GB
Mac Studio M4 Max · Macs
Llama 3.3 70B Instruct
Q4_K_M · ~8 tok/s
51.3 GB · up to 128Kfaster
Mistral Large Instruct 2407
Q4_K_M · ~5 tok/s
86.4 GB · up to 32K
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
Llama 3.3 70B Instruct
Mistral Large Instruct 2407
Coding
Llama 3.3 70B Instruct
Mistral Large Instruct 2407
no data
Science
Llama 3.3 70B Instruct
Mistral Large Instruct 2407
no data
Reasoning
Llama 3.3 70B Instruct
Mistral Large Instruct 2407
Llama 3.3 70B Instruct · #84 of 100 open models · #34 of 49 that fit ≤64 GB
Mistral Large Instruct 2407 · #81 of 100 open models · #42 of 60 that fit ≤96 GB
Benchmarks
7 shared boards: Llama 3.3 70B Instruct leads 1, Mistral Large Instruct 2407 leads 5 · Benchmarks imported Oct 2026
Llama 3.3 70B Instruct
47.4%#60/98 open
IndependentEpoch AI · Dec 2024
Mistral Large Instruct 2407
49.0%#56/98 open
IndependentEpoch AI · Jul 2024
Llama 3.3 70B Instruct
5.1%#68/86 open
IndependentEpoch AI · Dec 2024
Mistral Large Instruct 2407
8.5%#59/86 open
IndependentEpoch AI · Jul 2024
Llama 3.3 70B Instruct
65.9%#44/145 open
IndependentTIGER-Lab
Mistral Large Instruct 2407
65.9%#45/145 open
IndependentTIGER-Lab
Llama 3.3 70B Instruct
19.9%#23/25 open
IndependentEpoch AI · Dec 2024
Mistral Large Instruct 2407
22.5%#21/25 open
IndependentEpoch AI · Jul 2024
Llama 3.3 70B Instruct
41.6%#20/39 open
IndependentEpoch AI · Dec 2024
Mistral Large Instruct 2407
44.8%#17/39 open
IndependentEpoch AI · Jul 2024
Llama 3.3 70B Instruct
59.5%#51/68 open
Lab-reportedEpoch AI · Dec 2024
Mistral Large Instruct 2407
61.2%#45/68 open
Lab-reportedEpoch AI · Jul 2024
Llama 3.3 70B Instruct
58.6#11/26 open
Lab-reportedEpoch AI · Dec 2024
Mistral Large Instruct 2407
57.1#20/26 open
Lab-reportedEpoch AI · Jul 2024
8 more boards (other shared and single-model boards)
Llama 3.3 70B Instruct
1318.1#95/187 open
ArenaLMArena · Oct 2026
Mistral Large Instruct 2407
1314.4#98/187 open
ArenaLMArena · Oct 2026
Llama 3.3 70B Instruct
86.3%#2/85 open
Lab-reportedEpoch AI · Dec 2024
Mistral Large Instruct 2407
80.0%#11/85 open
Lab-reportedEpoch AI · Jul 2024
VRAM by quantization
Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.
Q2_K
Llama 3.3 70B Instruct
33.0 GB26.4 GB file
32K 43.1 GB · 128K 75.5 GB
Mistral Large Instruct 2407
57.3 GB45.2 GB file
32K 69.1 GB · 128K 107 GB
Q3_K_S
Llama 3.3 70B Instruct
34.0 GB30.9 GB file
32K 44.1 GB · 128K 76.4 GB
Mistral Large Instruct 2407
59.0 GB52.8 GB file
32K 70.8 GB · 128K 109 GB
Q3_K_M
Llama 3.3 70B Instruct
37.8 GB34.3 GB file
32K 48.0 GB · 128K 80.3 GB
Mistral Large Instruct 2407
65.8 GB59.1 GB file
32K 77.6 GB · 128K 115 GB
Q4_0
Llama 3.3 70B Instruct
38.8 GB40.1 GB file
32K 48.9 GB · 128K 81.3 GB
Mistral Large Instruct 2407
67.4 GB69.1 GB file
32K 79.3 GB · 128K 117 GB
Q4_K_M
Llama 3.3 70B Instruct
46.6 GB42.5 GB file
32K 56.7 GB · 128K 89.1 GB
Mistral Large Instruct 2407
80.9 GB73.2 GB file
32K 92.7 GB · 128K 131 GB
Q5_K_M
Llama 3.3 70B Instruct
55.3 GB49.9 GB file
32K 65.4 GB · 128K 97.8 GB
Mistral Large Instruct 2407
96.1 GB86.5 GB file
32K 108 GB · 128K 146 GB
Q6_K
Llama 3.3 70B Instruct
64.0 GB57.9 GB file
32K 74.2 GB · 128K 107 GB
Mistral Large Instruct 2407
111 GB100.6 GB file
32K 123 GB · 128K 161 GB
Q8_0
Llama 3.3 70B Instruct
77.6 GB75.0 GB file
32K 87.7 GB · 128K 120 GB
Mistral Large Instruct 2407
135 GB130.3 GB file
32K 147 GB · 128K 184 GB
19 more quants
IQ2_XXS
Llama 3.3 70B Instruct
21.3 GB19.4 GB file
32K 31.5 GB · 128K 63.8 GB
Mistral Large Instruct 2407
37.1 GB32.4 GB file
32K 48.9 GB · 128K 86.7 GB
IQ2_XS
Llama 3.3 70B Instruct
23.3 GB21.1 GB file
32K 33.4 GB · 128K 65.8 GB
Mistral Large Instruct 2407
40.5 GB36.1 GB file
32K 52.3 GB · 128K 90.1 GB
IQ2_S
Llama 3.3 70B Instruct
24.3 GB22.2 GB file
32K 34.4 GB · 128K 66.7 GB
Mistral Large Instruct 2407
42.2 GB38.4 GB file
32K 54.0 GB · 128K 91.8 GB
IQ2_M
Llama 3.3 70B Instruct
26.2 GB24.3 GB file
32K 36.3 GB · 128K 68.7 GB
Mistral Large Instruct 2407
45.5 GB41.6 GB file
32K 57.3 GB · 128K 95.1 GB
IQ3_XXS
Llama 3.3 70B Instruct
30.1 GB27.7 GB file
32K 40.2 GB · 128K 72.6 GB
Mistral Large Instruct 2407
52.3 GB47.0 GB file
32K 64.1 GB · 128K 102 GB
Q2_K_S
Llama 3.3 70B Instruct
31.0 GB24.5 GB file
32K 41.2 GB · 128K 73.5 GB
Mistral Large Instruct 2407
54.0 GB41.6 GB file
32K 65.8 GB · 128K 104 GB
IQ3_XS
Llama 3.3 70B Instruct
32.0 GB29.3 GB file
32K 42.1 GB · 128K 74.5 GB
Mistral Large Instruct 2407
55.6 GB50.1 GB file
32K 67.4 GB · 128K 105 GB
IQ3_S
Llama 3.3 70B Instruct
33.0 GB30.9 GB file
32K 43.1 GB · 128K 75.5 GB
Mistral Large Instruct 2407
not listed
IQ3_M
Llama 3.3 70B Instruct
34.9 GB31.9 GB file
32K 45.0 GB · 128K 77.4 GB
Mistral Large Instruct 2407
60.7 GB55.3 GB file
32K 72.5 GB · 128K 110 GB
Q3_K_L
Llama 3.3 70B Instruct
39.8 GB37.1 GB file
32K 49.9 GB · 128K 82.3 GB
Mistral Large Instruct 2407
69.1 GB64.6 GB file
32K 80.9 GB · 128K 119 GB
IQ4_XS
Llama 3.3 70B Instruct
41.7 GB37.9 GB file
32K 51.8 GB · 128K 84.2 GB
Mistral Large Instruct 2407
72.5 GB65.4 GB file
32K 84.3 GB · 128K 122 GB
IQ4_NL
Llama 3.3 70B Instruct
43.7 GB40.1 GB file
32K 53.8 GB · 128K 86.2 GB
Mistral Large Instruct 2407
not listed
Q4_1
Llama 3.3 70B Instruct
43.7 GB44.3 GB file
32K 53.8 GB · 128K 86.2 GB
Mistral Large Instruct 2407
not listed
Q4_K_S
Llama 3.3 70B Instruct
43.7 GB40.3 GB file
32K 53.8 GB · 128K 86.2 GB
Mistral Large Instruct 2407
75.9 GB69.6 GB file
32K 87.7 GB · 128K 125 GB
Q4_K_L
Llama 3.3 70B Instruct
47.5 GB43.3 GB file
32K 57.7 GB · 128K 90.0 GB
Mistral Large Instruct 2407
not listed
Q5_0
Llama 3.3 70B Instruct
48.5 GB48.7 GB file
32K 58.6 GB · 128K 91.0 GB
Mistral Large Instruct 2407
84.3 GB84.4 GB file
32K 96.1 GB · 128K 134 GB
Q5_K_S
Llama 3.3 70B Instruct
53.4 GB48.7 GB file
32K 63.5 GB · 128K 95.9 GB
Mistral Large Instruct 2407
92.7 GB84.4 GB file
32K 105 GB · 128K 142 GB
Q5_K_L
Llama 3.3 70B Instruct
56.3 GB50.6 GB file
32K 66.4 GB · 128K 98.8 GB
Mistral Large Instruct 2407
not listed
BF16
Llama 3.3 70B Instruct
155 GB141.1 GB file
32K 165 GB · 128K 198 GB
Mistral Large Instruct 2407
270 GB~245.2 GB file
32K 282 GB · 128K 319 GB
Llama 3.3 70B Instruct
Ollama · Q4_K_M
ollama run hf.co/MaziyarPanahi/Llama-3.3-70B-Instruct-GGUF:Q4_K_MMistral Large Instruct 2407
Ollama · Q4_K_M
ollama run hf.co/MaziyarPanahi/Mistral-Large-Instruct-2407-GGUF:Q4_K_M| Spec | Llama 3.3 70B Instruct | Mistral Large Instruct 2407 |
|---|---|---|
| Publisher | Meta | Mistral AI |
| Parameters | 70.6B | 122.6B |
| Context | 128K tokens | — |
| Modalities | Chat, Function-calling | Chat, Function-calling |
| License | llama3.3 | Other |
| Released | Nov 2024 | Jul 2024 |
Frequently Asked Questions
Which is better, Llama 3.3 70B Instruct or Mistral Large Instruct 2407?
Llama 3.3 70B Instruct and Mistral Large Instruct 2407 match in llmrun Score, with values of 79.7 versus 81.1; neither fits 24 GB. Choose Llama 3.3 70B Instruct for fits a 48 GB GPU and fits a Mac 64 GB; choose Mistral Large Instruct 2407 for ahead on the benchmarks.
Which needs less VRAM, Llama 3.3 70B Instruct or Mistral Large Instruct 2407?
At Q4_K_M, Llama 3.3 70B Instruct requires 46.6 GB while Mistral Large Instruct 2407 needs 80.9 GB, making Llama 3.3 70B Instruct the easier model to run on your local setup.
Can I run Llama 3.3 70B Instruct or Mistral Large Instruct 2407 on a 16 GB GPU?
Llama 3.3 70B Instruct does not fit (smallest quant needs 26.1 GB); Mistral Large Instruct 2407 does not fit (smallest quant needs 42.6 GB).