C4ai Command A 03 2025 vs Llama 3.3 70B Instruct
C4ai Command A 03 2025 and Llama 3.3 70B Instruct match in llmrun Score, with values of 79.8 versus 79.7; neither fits 24 GB.
Model B · Meta
Llama 3.3 70B Instruct70.6B · 128K context · llama3.3 · Nov 2024
Choose C4ai Command A 03 2025 if you want…
Newer release
Mar 2025 vs Nov 2024
Choose Llama 3.3 70B Instruct if you want…
Fits a 48 GB GPU
Q4_K_M · 51.3 GB; C4ai Command A 03 2025 needs 39 GB at its smallest quant
Lighter on memory
46.6 vs 73.3 GB at Q4_K_M
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
C4ai Command A 03 2025
Doesn't fit
smallest quant needs 39 GB
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
12 GB
RTX 3060 12GB · 12 GB tier
C4ai Command A 03 2025
Doesn't fit
smallest quant needs 39 GB
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
16 GB
RTX 5060 Ti 16GB · 16 GB tier
C4ai Command A 03 2025
Doesn't fit
smallest quant needs 39 GB
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
24 GBreference
RTX 4090 · 24 GB tier
C4ai Command A 03 2025
Doesn't fit
smallest quant needs 39 GB
Llama 3.3 70B Instruct
Doesn't fit
smallest quant needs 26.1 GB
32 GB
RTX 5090 · 32 GB tier
C4ai Command A 03 2025
Doesn't fit
smallest quant needs 39 GB
Llama 3.3 70B Instruct
IQ2_M · ~45 tok/s
30.9 GB · up to 16Kfits
48 GB
RTX A6000 · 48 GB tier
C4ai Command A 03 2025
IQ2_M · ~12 tok/s
46.6 GB · up to 16Kmore headroom
Llama 3.3 70B Instruct
Q4_K_M · ~11 tok/stight
51.3 GB · up to 4K
Mac 32 GB
Mac Mini M4 · Macs
C4ai Command A 03 2025
Doesn't fit
smallest quant needs 39 GB
Llama 3.3 70B Instruct
IQ2_XS · ~4 tok/s
28 GB · up to 16Kfits
Mac 64 GB
Mac Studio M4 Max · Macs
C4ai Command A 03 2025
IQ3_M · ~7 tok/s
60.3 GB · up to 16K
Llama 3.3 70B Instruct
Q4_K_M · ~8 tok/s
51.3 GB · up to 32Kmore headroom
Mac 128 GB
Mac Studio M4 Max · Macs
C4ai Command A 03 2025
Q4_K_M · ~5 tok/s
78.7 GB · up to 32K
Llama 3.3 70B Instruct
Q4_K_M · ~8 tok/s
51.3 GB · up to 128Kfaster
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
C4ai Command A 03 2025
Llama 3.3 70B Instruct
Coding
C4ai Command A 03 2025
no data
Llama 3.3 70B Instruct
Science
C4ai Command A 03 2025
no data
Llama 3.3 70B Instruct
Reasoning
C4ai Command A 03 2025
Llama 3.3 70B Instruct
C4ai Command A 03 2025 · #83 of 100 open models · #44 of 60 that fit ≤96 GB
Llama 3.3 70B Instruct · #84 of 100 open models · #34 of 49 that fit ≤64 GB
Benchmarks
2 shared boards: C4ai Command A 03 2025 leads 1, Llama 3.3 70B Instruct leads 1 · Benchmarks imported Oct 2026
14 more boards (other shared and single-model boards)
C4ai Command A 03 2025
1353.7#73/187 open
ArenaLMArena · Oct 2026
Llama 3.3 70B Instruct
1318.1#95/187 open
ArenaLMArena · Oct 2026
Only C4ai Command A 03 2025 (1)
- Aider Polyglot12.0%
Only Llama 3.3 70B Instruct (12)
- GPQA Diamond47.4%
- AIME 2024/20255.1%
- MMLU-Pro65.9%
- SciCode26.0%
- SimpleBench19.9%
- MATH Level 541.6%
- MMLU86.3%
- WeirdML14.4%
- BALROG23.0%
- CritPt0.0%
- Fiction.LiveBench33.3%
- ForecastBench58.6
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
C4ai Command A 03 2025
51.9 GB42.1 GB file
32K 63.4 GB · 128K 100 GB
Llama 3.3 70B Instruct
33.0 GB26.4 GB file
32K 43.1 GB · 128K 75.5 GB
Q3_K_S
C4ai Command A 03 2025
53.5 GB49.0 GB file
32K 64.9 GB · 128K 102 GB
Llama 3.3 70B Instruct
34.0 GB30.9 GB file
32K 44.1 GB · 128K 76.4 GB
Q3_K_M
C4ai Command A 03 2025
59.6 GB54.4 GB file
32K 71.0 GB · 128K 108 GB
Llama 3.3 70B Instruct
37.8 GB34.3 GB file
32K 48.0 GB · 128K 80.3 GB
Q4_0
C4ai Command A 03 2025
61.1 GB63.5 GB file
32K 72.6 GB · 128K 109 GB
Llama 3.3 70B Instruct
38.8 GB40.1 GB file
32K 48.9 GB · 128K 81.3 GB
Q4_K_M
C4ai Command A 03 2025
73.3 GB67.1 GB file
32K 84.8 GB · 128K 122 GB
Llama 3.3 70B Instruct
46.6 GB42.5 GB file
32K 56.7 GB · 128K 89.1 GB
Q5_K_M
C4ai Command A 03 2025
87.0 GB78.8 GB file
32K 98.5 GB · 128K 135 GB
Llama 3.3 70B Instruct
55.3 GB49.9 GB file
32K 65.4 GB · 128K 97.8 GB
Q6_K
C4ai Command A 03 2025
101 GB91.1 GB file
32K 112 GB · 128K 149 GB
Llama 3.3 70B Instruct
64.0 GB57.9 GB file
32K 74.2 GB · 128K 107 GB
Q8_0
C4ai Command A 03 2025
122 GB118.0 GB file
32K 134 GB · 128K 170 GB
Llama 3.3 70B Instruct
77.6 GB75.0 GB file
32K 87.7 GB · 128K 120 GB
19 more quants
IQ2_XXS
C4ai Command A 03 2025
33.6 GB30.5 GB file
32K 45.1 GB · 128K 81.8 GB
Llama 3.3 70B Instruct
21.3 GB19.4 GB file
32K 31.5 GB · 128K 63.8 GB
IQ2_XS
C4ai Command A 03 2025
36.7 GB33.7 GB file
32K 48.1 GB · 128K 84.9 GB
Llama 3.3 70B Instruct
23.3 GB21.1 GB file
32K 33.4 GB · 128K 65.8 GB
IQ2_S
C4ai Command A 03 2025
38.2 GB35.5 GB file
32K 49.7 GB · 128K 86.4 GB
Llama 3.3 70B Instruct
24.3 GB22.2 GB file
32K 34.4 GB · 128K 66.7 GB
IQ2_M
C4ai Command A 03 2025
41.2 GB38.4 GB file
32K 52.7 GB · 128K 89.5 GB
Llama 3.3 70B Instruct
26.2 GB24.3 GB file
32K 36.3 GB · 128K 68.7 GB
IQ3_XXS
C4ai Command A 03 2025
47.3 GB43.4 GB file
32K 58.8 GB · 128K 95.6 GB
Llama 3.3 70B Instruct
30.1 GB27.7 GB file
32K 40.2 GB · 128K 72.6 GB
Q2_K_S
C4ai Command A 03 2025
48.9 GB39.0 GB file
32K 60.4 GB · 128K 97.1 GB
Llama 3.3 70B Instruct
31.0 GB24.5 GB file
32K 41.2 GB · 128K 73.5 GB
IQ3_XS
C4ai Command A 03 2025
50.4 GB46.5 GB file
32K 61.9 GB · 128K 98.6 GB
Llama 3.3 70B Instruct
32.0 GB29.3 GB file
32K 42.1 GB · 128K 74.5 GB
IQ3_S
C4ai Command A 03 2025
51.9 GB49.1 GB file
32K 63.4 GB · 128K 100 GB
Llama 3.3 70B Instruct
33.0 GB30.9 GB file
32K 43.1 GB · 128K 75.5 GB
IQ3_M
C4ai Command A 03 2025
55.0 GB50.8 GB file
32K 66.5 GB · 128K 103 GB
Llama 3.3 70B Instruct
34.9 GB31.9 GB file
32K 45.0 GB · 128K 77.4 GB
Q3_K_L
C4ai Command A 03 2025
62.6 GB59.1 GB file
32K 74.1 GB · 128K 111 GB
Llama 3.3 70B Instruct
39.8 GB37.1 GB file
32K 49.9 GB · 128K 82.3 GB
IQ4_XS
C4ai Command A 03 2025
65.7 GB60.0 GB file
32K 77.2 GB · 128K 114 GB
Llama 3.3 70B Instruct
41.7 GB37.9 GB file
32K 51.8 GB · 128K 84.2 GB
IQ4_NL
C4ai Command A 03 2025
68.7 GB63.4 GB file
32K 80.2 GB · 128K 117 GB
Llama 3.3 70B Instruct
43.7 GB40.1 GB file
32K 53.8 GB · 128K 86.2 GB
Q4_1
C4ai Command A 03 2025
68.7 GB70.0 GB file
32K 80.2 GB · 128K 117 GB
Llama 3.3 70B Instruct
43.7 GB44.3 GB file
32K 53.8 GB · 128K 86.2 GB
Q4_K_S
C4ai Command A 03 2025
68.7 GB63.8 GB file
32K 80.2 GB · 128K 117 GB
Llama 3.3 70B Instruct
43.7 GB40.3 GB file
32K 53.8 GB · 128K 86.2 GB
Q4_K_L
C4ai Command A 03 2025
74.8 GB67.9 GB file
32K 86.3 GB · 128K 123 GB
Llama 3.3 70B Instruct
47.5 GB43.3 GB file
32K 57.7 GB · 128K 90.0 GB
Q5_0
C4ai Command A 03 2025
not listed
Llama 3.3 70B Instruct
48.5 GB48.7 GB file
32K 58.6 GB · 128K 91.0 GB
Q5_K_S
C4ai Command A 03 2025
84.0 GB76.8 GB file
32K 95.5 GB · 128K 132 GB
Llama 3.3 70B Instruct
53.4 GB48.7 GB file
32K 63.5 GB · 128K 95.9 GB
Q5_K_L
C4ai Command A 03 2025
not listed
Llama 3.3 70B Instruct
56.3 GB50.6 GB file
32K 66.4 GB · 128K 98.8 GB
BF16
C4ai Command A 03 2025
244 GB222.1 GB file
32K 256 GB · 128K 293 GB
Llama 3.3 70B Instruct
155 GB141.1 GB file
32K 165 GB · 128K 198 GB
C4ai Command A 03 2025
Ollama · Q4_K_M
ollama run hf.co/bartowski/CohereForAI_c4ai-command-a-03-2025-GGUF:Q4_K_MLlama 3.3 70B Instruct
Ollama · Q4_K_M
ollama run hf.co/MaziyarPanahi/Llama-3.3-70B-Instruct-GGUF:Q4_K_M| Spec | C4ai Command A 03 2025 | Llama 3.3 70B Instruct |
|---|---|---|
| Publisher | Cohere | Meta |
| Parameters | 111.1B | 70.6B |
| Context | — | 128K tokens |
| Modalities | Chat, Function-calling | Chat, Function-calling |
| License | CC BY-NC 4.0 | llama3.3 |
| Released | Mar 2025 | Nov 2024 |
Frequently Asked Questions
Which is better, C4ai Command A 03 2025 or Llama 3.3 70B Instruct?
C4ai Command A 03 2025 and Llama 3.3 70B Instruct match in llmrun Score, with values of 79.8 versus 79.7; neither fits 24 GB. Choose C4ai Command A 03 2025 for newer release; choose Llama 3.3 70B Instruct for fits a 48 GB GPU and lighter on memory.
Which needs less VRAM, C4ai Command A 03 2025 or Llama 3.3 70B Instruct?
At Q4_K_M, C4ai Command A 03 2025 requires 73.3 GB while Llama 3.3 70B Instruct needs 46.6 GB, making Llama 3.3 70B Instruct the easier model to run on your local setup.
Can I run C4ai Command A 03 2025 or Llama 3.3 70B Instruct on a 16 GB GPU?
C4ai Command A 03 2025 does not fit (smallest quant needs 39 GB); Llama 3.3 70B Instruct does not fit (smallest quant needs 26.1 GB).