Meta Llama 3 8B Instruct vs Qwen3.5 9B
Qwen3.5 9B leads llmrun Score 99.4 to 62.0; both fit 24 GB at Q4_K_M.
Model B · Alibaba
Qwen3.5 9B9.7B · 256K context · Apache 2.0 · Feb 2026
Choose Meta Llama 3 8B Instruct if you want…
Lighter on memory
5.3 vs 6.4 GB at Q4_K_M
Also does chat
Qwen3.5 9B doesn't list it
Choose Qwen3.5 9B if you want…
Higher llmrun Score
99.4 vs 62.0
Stronger at reasoning
97.0 vs 65.2
Ahead on the benchmarks
4 of 4 shared boards
Also does vision, function-calling
Meta Llama 3 8B Instruct doesn't list it
More permissive license
Apache 2.0 vs Llama 3 Community
Newer release
Feb 2026 vs Apr 2024
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
Meta Llama 3 8B Instruct
Q4_K_M · ~33 tok/s
7.9 GB · up to 16Kfaster
Qwen3.5 9B
Q4_K_M · ~28 tok/s
6.6 GB · up to 32K
12 GB
RTX 3060 12GB · 12 GB tier
Meta Llama 3 8B Instruct
Q4_K_M · ~44 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~37 tok/s
6.6 GB · up to 128K
16 GB
RTX 5060 Ti 16GB · 16 GB tier
Meta Llama 3 8B Instruct
Q4_K_M · ~55 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~46 tok/s
6.6 GB · up to 256K
24 GBreference
RTX 4090 · 24 GB tier
Meta Llama 3 8B Instruct
Q4_K_M · ~124 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~103 tok/s
6.6 GB · up to 256K
32 GB
RTX 5090 · 32 GB tier
Meta Llama 3 8B Instruct
Q4_K_M · ~220 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~183 tok/s
6.6 GB · up to 256K
48 GB
RTX A6000 · 48 GB tier
Meta Llama 3 8B Instruct
Q4_K_M · ~94 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~79 tok/s
6.6 GB · up to 256K
Mac 32 GB
Mac Mini M4 · Macs
Meta Llama 3 8B Instruct
Q4_K_M · ~16 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~13 tok/s
6.6 GB · up to 256K
Mac 64 GB
Mac Studio M4 Max · Macs
Meta Llama 3 8B Instruct
Q4_K_M · ~72 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~60 tok/s
6.6 GB · up to 256K
Mac 128 GB
Mac Studio M4 Max · Macs
Meta Llama 3 8B Instruct
Q4_K_M · ~72 tok/s
7.9 GB · up to 32Kfaster
Qwen3.5 9B
Q4_K_M · ~60 tok/s
6.6 GB · up to 256K
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
Meta Llama 3 8B Instruct
Qwen3.5 9B
Science
Meta Llama 3 8B Instruct
no data
Qwen3.5 9B
Reasoning
Meta Llama 3 8B Instruct
Qwen3.5 9B
Meta Llama 3 8B Instruct · #105 of 106 open models · #36 of 36 that fit 24 GB
Qwen3.5 9B · #59 of 106 open models · #10 of 36 that fit 24 GB
Benchmarks
4 shared boards: Meta Llama 3 8B Instruct leads 0, Qwen3.5 9B leads 4 · Benchmarks imported Nov 2026
Meta Llama 3 8B Instruct
26.1%#91/100 open
IndependentEpoch AI · Apr 2024
Qwen3.5 9B
79.0%#28/100 open
IndependentEpoch AI · Feb 2026
Meta Llama 3 8B Instruct
1.9%#78/88 open
IndependentEpoch AI · Apr 2024
Qwen3.5 9B
61.7%#39/88 open
IndependentEpoch AI · Feb 2026
Meta Llama 3 8B Instruct
41.0%#100/146 open
IndependentTIGER-Lab
Qwen3.5 9B
82.5%#17/146 open
IndependentTIGER-Lab
Meta Llama 3 8B Instruct
43.9%#68/70 open
Lab-reportedEpoch AI · Apr 2024
Qwen3.5 9B
71.2%#33/70 open
Lab-reportedEpoch AI · Feb 2026
12 more boards (other shared and single-model boards)
Only Meta Llama 3 8B Instruct (8)
- LMArena Text1223.7
- MATH Level 56.1%
- MMLU68.8%
- ForecastBench59.4
- WinoGrande65.0%
- ARC Challenge82.8%
- OpenBookQA82.6%
- TriviaQA67.7%
Only Qwen3.5 9B (4)
- Terminal-Bench9.2%
- SciCode29.5%
- CritPt0.6%
- LMCA24.5%
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
Meta Llama 3 8B Instruct
3.8 GB3.2 GB file
32K 9.3 GB · 128K 26.9 GB
Qwen3.5 9B
4.7 GB4.3 GB file
32K 5.5 GB · 128K 8.7 GB
Q3_K_S
Meta Llama 3 8B Instruct
3.9 GB3.7 GB file
32K 9.4 GB · 128K 27.0 GB
Qwen3.5 9B
4.8 GB4.3 GB file
32K 5.6 GB · 128K 8.8 GB
Q3_K_M
Meta Llama 3 8B Instruct
4.3 GB4.0 GB file
32K 9.8 GB · 128K 27.4 GB
Qwen3.5 9B
5.3 GB4.7 GB file
32K 6.1 GB · 128K 9.3 GB
Q4_0
Meta Llama 3 8B Instruct
4.4 GB4.7 GB file
32K 9.9 GB · 128K 27.5 GB
Qwen3.5 9B
5.4 GB5.4 GB file
32K 6.2 GB · 128K 9.4 GB
Q4_K_M
Meta Llama 3 8B Instruct
5.3 GB4.9 GB file
32K 10.8 GB · 128K 28.4 GB
Qwen3.5 9B
6.4 GB5.7 GB file
32K 7.2 GB · 128K 10.4 GB
Q5_K_M
Meta Llama 3 8B Instruct
6.3 GB5.7 GB file
32K 11.8 GB · 128K 29.4 GB
Qwen3.5 9B
7.5 GB6.6 GB file
32K 8.3 GB · 128K 11.5 GB
Q6_K
Meta Llama 3 8B Instruct
7.3 GB6.6 GB file
32K 12.8 GB · 128K 30.4 GB
Qwen3.5 9B
8.5 GB7.5 GB file
32K 9.3 GB · 128K 12.6 GB
Q8_0
Meta Llama 3 8B Instruct
8.8 GB8.5 GB file
32K 14.3 GB · 128K 32.0 GB
Qwen3.5 9B
10.2 GB9.5 GB file
32K 11.0 GB · 128K 14.3 GB
20 more quants
IQ2_XXS
Meta Llama 3 8B Instruct
2.4 GB2.4 GB file
32K 7.9 GB · 128K 25.6 GB
Qwen3.5 9B
3.2 GB3.2 GB file
32K 4.0 GB · 128K 7.3 GB
IQ2_XS
Meta Llama 3 8B Instruct
2.7 GB2.6 GB file
32K 8.2 GB · 128K 25.8 GB
Qwen3.5 9B
not listed
IQ2_S
Meta Llama 3 8B Instruct
2.8 GB2.8 GB file
32K 8.3 GB · 128K 25.9 GB
Qwen3.5 9B
3.6 GB3.5 GB file
32K 4.4 GB · 128K 7.6 GB
IQ2_M
Meta Llama 3 8B Instruct
3.0 GB2.9 GB file
32K 8.5 GB · 128K 26.1 GB
Qwen3.5 9B
3.8 GB3.6 GB file
32K 4.6 GB · 128K 7.9 GB
IQ3_XXS
Meta Llama 3 8B Instruct
3.4 GB3.3 GB file
32K 8.9 GB · 128K 26.5 GB
Qwen3.5 9B
4.3 GB4.0 GB file
32K 5.1 GB · 128K 8.3 GB
Q2_K_S
Meta Llama 3 8B Instruct
3.5 GB3.0 GB file
32K 9.0 GB · 128K 26.7 GB
Qwen3.5 9B
not listed
IQ3_XS
Meta Llama 3 8B Instruct
3.6 GB3.5 GB file
32K 9.1 GB · 128K 26.8 GB
Qwen3.5 9B
4.6 GB4.8 GB file
32K 5.4 GB · 128K 8.6 GB
IQ3_S
Meta Llama 3 8B Instruct
3.8 GB3.7 GB file
32K 9.3 GB · 128K 26.9 GB
Qwen3.5 9B
4.7 GB3.2 GB file
32K 5.5 GB · 128K 8.7 GB
IQ3_M
Meta Llama 3 8B Instruct
4.0 GB3.8 GB file
32K 9.5 GB · 128K 27.1 GB
Qwen3.5 9B
4.9 GB5.0 GB file
32K 5.7 GB · 128K 8.9 GB
Q3_K_L
Meta Llama 3 8B Instruct
4.5 GB4.3 GB file
32K 10.0 GB · 128K 27.7 GB
Qwen3.5 9B
5.5 GB5.4 GB file
32K 6.3 GB · 128K 9.6 GB
IQ4_XS
Meta Llama 3 8B Instruct
4.8 GB4.4 GB file
32K 10.3 GB · 128K 27.9 GB
Qwen3.5 9B
5.8 GB5.2 GB file
32K 6.6 GB · 128K 9.8 GB
IQ4_NL
Meta Llama 3 8B Instruct
5.0 GB4.7 GB file
32K 10.5 GB · 128K 28.1 GB
Qwen3.5 9B
6.0 GB5.4 GB file
32K 6.8 GB · 128K 10.0 GB
Q4_1
Meta Llama 3 8B Instruct
5.0 GB5.1 GB file
32K 10.5 GB · 128K 28.1 GB
Qwen3.5 9B
6.0 GB5.8 GB file
32K 6.8 GB · 128K 10.0 GB
Q4_K_S
Meta Llama 3 8B Instruct
5.0 GB4.7 GB file
32K 10.5 GB · 128K 28.1 GB
Qwen3.5 9B
6.0 GB5.4 GB file
32K 6.8 GB · 128K 10.0 GB
Q4_K_L
Meta Llama 3 8B Instruct
not listed
Qwen3.5 9B
6.5 GB6.9 GB file
32K 7.3 GB · 128K 10.5 GB
Q5_0
Meta Llama 3 8B Instruct
5.5 GB5.6 GB file
32K 11.0 GB · 128K 28.6 GB
Qwen3.5 9B
not listed
Q5_1
Meta Llama 3 8B Instruct
6.1 GB6.1 GB file
32K 11.6 GB · 128K 29.2 GB
Qwen3.5 9B
not listed
Q5_K_S
Meta Llama 3 8B Instruct
6.1 GB5.6 GB file
32K 11.6 GB · 128K 29.2 GB
Qwen3.5 9B
7.2 GB6.4 GB file
32K 8.0 GB · 128K 11.2 GB
Q5_K_L
Meta Llama 3 8B Instruct
not listed
Qwen3.5 9B
7.6 GB7.7 GB file
32K 8.4 GB · 128K 11.6 GB
BF16
Meta Llama 3 8B Instruct
17.7 GB~16.1 GB file
32K 23.2 GB · 128K 40.8 GB
Qwen3.5 9B
19.9 GB17.9 GB file
32K 20.7 GB · 128K 23.9 GB
Meta Llama 3 8B Instruct
Ollama · Q4_K_M
ollama run hf.co/MaziyarPanahi/Meta-Llama-3-8B-Instruct-GGUF:Q4_K_MQwen3.5 9B
Ollama · Q4_K_M
ollama run hf.co/unsloth/Qwen3.5-9B-GGUF:Q4_K_M| Spec | Meta Llama 3 8B Instruct | Qwen3.5 9B |
|---|---|---|
| Publisher | Meta | Alibaba |
| Parameters | 8.0B | 9.7B |
| Architecture | — | Qwen3_5ForConditionalGeneration |
| Context | — | 256K tokens |
| Modalities | Chat | Vision, Function-calling |
| License | Llama 3 Community | Apache 2.0 |
| Released | Apr 2024 | Feb 2026 |
Frequently Asked Questions
Which is better, Meta Llama 3 8B Instruct or Qwen3.5 9B?
Qwen3.5 9B leads llmrun Score 99.4 to 62.0; both fit 24 GB at Q4_K_M. Choose Meta Llama 3 8B Instruct for lighter on memory and also does chat; choose Qwen3.5 9B for higher llmrun Score and stronger at reasoning.
Which needs less VRAM, Meta Llama 3 8B Instruct or Qwen3.5 9B?
At Q4_K_M, Meta Llama 3 8B Instruct requires 5.3 GB while Qwen3.5 9B needs 6.4 GB, making Meta Llama 3 8B Instruct the easier model to run on your local setup.
Which is faster on RTX 4090?
Estimated decode speed on RTX 4090 is about 124 tok/s for Meta Llama 3 8B Instruct (Q4_K_M) and 103 tok/s for Qwen3.5 9B (Q4_K_M), so Meta Llama 3 8B Instruct is faster. Estimates, ±20%.
Can I run Meta Llama 3 8B Instruct or Qwen3.5 9B on a 16 GB GPU?
Meta Llama 3 8B Instruct fits at Q4_K_M; Qwen3.5 9B fits at Q4_K_M.