Your rig
Radeon RX 7600 XT · 16 GB
Which one should I actually run?
Best models for your Radeon RX 7600 XT
What do you want to do? Each row is the best-ranked model on that use case's shortlist that fits at Q3_K_M and 8K. The ordering is editorial; the sizes and speeds are computed. How the shortlists work.
| Recommendation | Model | Why | Quant · total | Tok/s est. | |
|---|---|---|---|---|---|
| Best overall | Qwen3.8 27B Alibaba · Apache 2.0 |
Best capability per gigabyte: a current-generation dense 27B with vision and 262K context. | Q3_K_M · 13.8 GB | 14 | Tight fit |
| Best for coding | Qwen3.8 27B Alibaba · Apache 2.0 |
Terminal-Bench 73, DeepSWE 42 — a generation ahead of anything else that fits 24 GB. | Q3_K_M · 13.8 GB | 14 | Tight fit |
| Best reasoning | Qwen3.8 27B Alibaba · Apache 2.0 |
Thinking mode plus the best agentic reasoning at 24 GB. | Q3_K_M · 13.8 GB | 14 | Tight fit |
| Best for writing | Gemma 4 12B Google · Apache 2.0 |
The 12 GB writing pick; 140+ languages. | Q3_K_M · 6.9 GB | 32 | Runs great |
| Best vision | Qwen3.8 27B Alibaba · Apache 2.0 |
Image and video in, OSWorld 84 — the strongest local vision-language model at 24 GB. | Q3_K_M · 13.8 GB | 14 | Tight fit |
| Best for agents | Qwen3.8 27B Alibaba · Apache 2.0 |
Computer use and tool calling are what 3.8 was trained for (OSWorld 84). | Q3_K_M · 13.8 GB | 14 | Tight fit |
| Best translation | Qwen3.8 27B Alibaba · Apache 2.0 |
119 languages, and Qwen translations read as idiomatic rather than literal. | Q3_K_M · 13.8 GB | 14 | Tight fit |
| Fastest good model | Gemma 4 26B-A4B Google · Apache 2.0 |
3.8B active and a small file — the quickest Gemma 4 with vision. | Q3_K_M · 13.2 GB | 39 | Tight fit |
| Best long context | Qwen3.5 9B Alibaba · Apache 2.0 |
262K native in 8–12 GB. | Q3_K_M · 9.0 GB at 128K context |
40 | Runs great |
Everything that fits
Runs great
38 of 80 · fits with headroom for a longer prompt| Model | Params | Quant | Weights | +KV 8K | Total | VRAM headroom | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| gpt-oss 20B | 20.9B MoE / 3.6B act | MXFP4 | 10.8 GB | 0.19 GB | 11.6 GB |
2.8 GB free
|
36 | Ollama | Run |
| Qwen3 14B | 14.8B | Q3_K_M | 6.7 GB | 1.25 GB | 8.6 GB |
5.8 GB free
|
26 | Ollama | Run |
| DeepSeek-R1-Distill-Qwen 14B | 14.8B | Q3_K_M | 6.7 GB | 1.50 GB | 8.8 GB |
5.6 GB free
|
26 | Ollama | Run |
| Qwen2.5-Coder 14B | 14.8B | Q3_K_M | 6.7 GB | 1.50 GB | 8.8 GB |
5.6 GB free
|
26 | Ollama | Run |
| Phi-4 14B | 14.7B | Q3_K_M | 6.7 GB | 1.56 GB | 8.9 GB |
5.5 GB free
|
26 | Ollama | Run |
| Ministral 3 14B | 13.9B | Q3_K_M | 6.3 GB | 1.25 GB | 8.2 GB |
6.2 GB free
|
28 | Ollama | Run |
| Gemma 3 12B | 12.2B | Q3_K_M | 5.6 GB | 0.81 GB | 7.0 GB |
7.4 GB free
|
31 | Ollama | Run |
| Mistral NeMo 12B | 12.2B | Q3_K_M | 5.6 GB | 1.25 GB | 7.4 GB |
7.0 GB free
|
31 | Ollama | Run |
| Gemma 4 12B | 12B | Q3_K_M | 5.5 GB | 0.81 GB | 6.9 GB |
7.5 GB free
|
32 | Ollama | Run |
| Qwen3.5 9B | 9.65B | Q3_K_M | 4.4 GB | 0.25 GB | 5.2 GB |
9.2 GB free
|
40 | Ollama | Run |
| Ornith 1.5 9B | 9.41B | Q3_K_M | 4.3 GB | 0.25 GB | 5.1 GB |
9.3 GB free
|
41 | Ollama | Run |
| Ministral 3 8B | 8.92B | Q3_K_M | 4.1 GB | 1.06 GB | 5.7 GB |
8.7 GB free
|
43 | Ollama | Run |
| Granite 4.1 8B | 8.79B | Q3_K_M | 4.0 GB | 1.25 GB | 5.9 GB |
8.5 GB free
|
44 | Ollama | Run |
| LFM2.5 8B-A1B | 8.47B MoE / 1.5B act | Q3_K_M | 3.9 GB | 0.09 GB | 4.5 GB |
9.9 GB free
|
98 | Ollama | Run |
| Fara 7B | 8.29B | Q3_K_M | 3.8 GB | 0.44 GB | 4.8 GB |
9.6 GB free
|
46 | llama.cpp | Run |
| Qwen3 8B | 8.19B | Q3_K_M | 3.7 GB | 1.13 GB | 5.5 GB |
8.9 GB free
|
47 | Ollama | Run |
| Llama 3.1 8B Instruct | 8.03B | Q3_K_M | 3.7 GB | 1.00 GB | 5.3 GB |
9.1 GB free
|
48 | Ollama | Run |
| Gemma 4 E4B | 8.0B | Q3_K_M | 3.6 GB | 0.14 GB | 4.4 GB |
10.0 GB free
|
48 | Ollama | Run |
| Ling 3.0 Tiny 7.9B-A1.3B | 7.9B MoE / 1.3B act | Q3_K_M | 3.6 GB | 0.05 GB | 4.2 GB |
10.2 GB free
|
113 | llama.cpp | Run |
| DeepSeek-R1-Distill-Qwen 7B | 7.62B | Q3_K_M | 3.5 GB | 0.44 GB | 4.5 GB |
9.9 GB free
|
50 | Ollama | Run |
| Qwen2.5-Coder 7B | 7.62B | Q3_K_M | 3.5 GB | 0.44 GB | 4.5 GB |
9.9 GB free
|
50 | Ollama | Run |
| Olmo 3 7B Instruct | 7.3B | Q3_K_M | 3.3 GB | 2.50 GB | 6.4 GB |
8.0 GB free
|
52 | Ollama | Run |
| Mistral 7B Instruct v0.3 | 7.25B | Q3_K_M | 3.3 GB | 1.00 GB | 4.9 GB |
9.5 GB free
|
53 | Ollama | Run |
| Gemma 4 E2B | 5.1B | Q3_K_M | 2.3 GB | 0.07 GB | 3.0 GB |
11.4 GB free
|
75 | Ollama | Run |
| Qwen3.5 4B | 4.66B | Q3_K_M | 2.1 GB | 0.25 GB | 3.0 GB |
11.4 GB free
|
82 | Ollama | Run |
| Gemma 3 4B | 4.3B | Q3_K_M | 2.0 GB | 0.30 GB | 2.9 GB |
11.5 GB free
|
89 | Ollama | Run |
| Qwen3 4B | 4.02B | Q3_K_M | 1.8 GB | 1.13 GB | 3.6 GB |
10.8 GB free
|
95 | Ollama | Run |
| Ministral 3 3B | 3.85B | Q3_K_M | 1.8 GB | 0.81 GB | 3.2 GB |
11.2 GB free
|
99 | Ollama | Run |
| Phi-4-mini 3.8B | 3.84B | Q3_K_M | 1.7 GB | 1.00 GB | 3.3 GB |
11.1 GB free
|
100 | Ollama | Run |
| Granite 4.1 3B | 3.4B | Q3_K_M | 1.5 GB | 0.63 GB | 2.8 GB |
11.6 GB free
|
113 | Ollama | Run |
| Llama 3.2 3B Instruct | 3.21B | Q3_K_M | 1.5 GB | 0.88 GB | 2.9 GB |
11.5 GB free
|
119 | Ollama | Run |
| LFM2.5 2.6B | 2.7B | Q3_K_M | 1.2 GB | 0.13 GB | 2.0 GB |
12.4 GB free
|
142 | llama.cpp | Run |
| Qwen3.5 2B | 2.27B | Q3_K_M | 1.0 GB | 0.09 GB | 1.7 GB |
12.7 GB free
|
169 | Ollama | Run |
| Qwen3 1.7B | 1.72B | Q3_K_M | 0.8 GB | 0.88 GB | 2.3 GB |
12.1 GB free
|
223 | Ollama | Run |
| Llama 3.2 1B Instruct | 1.24B | Q3_K_M | 0.6 GB | 0.25 GB | 1.4 GB |
13.0 GB free
|
309 | Ollama | Run |
| Gemma 3 1B | 1.0B | Q3_K_M | 0.5 GB | 0.04 GB | 1.1 GB |
13.3 GB free
|
383 | Ollama | Run |
| Qwen3.5 0.8B | 0.87B | Q3_K_M | 0.4 GB | 0.09 GB | 1.1 GB |
13.3 GB free
|
440 | Ollama | Run |
| Qwen3 0.6B | 0.6B | Q3_K_M | 0.3 GB | 0.88 GB | 1.7 GB |
12.7 GB free
|
638 | Ollama | Run |
Runs — tight
7 models · fits at 8K, drop context or quant to go longer| Model | Params | Quant | Weights | +KV 8K | Total | VRAM headroom | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| Muse Glimmer 30B | 29.8B | Q3_K_M | 13.6 GB | 0.18 GB | 14.3 GB |
0.1 GB free
|
13 | llama.cpp | Run |
| Qwen3.8 27B | 27.8B | Q3_K_M | 12.7 GB | 0.50 GB | 13.8 GB |
0.6 GB free
|
14 | llama.cpp | Run |
| Qwen3.6 27B | 27.8B | Q3_K_M | 12.7 GB | 0.50 GB | 13.8 GB |
0.6 GB free
|
14 | llama.cpp | Run |
| Gemma 3 27B | 27.4B | Q3_K_M | 12.5 GB | 1.03 GB | 14.1 GB |
0.3 GB free
|
14 | llama.cpp | Run |
| Gemma 4 26B-A4B | 26.5B MoE / 3.8B act | Q3_K_M | 12.1 GB | 0.51 GB | 13.2 GB |
1.2 GB free
|
39 | llama.cpp | Run |
| Devstral Small 2 24B | 24B | Q3_K_M | 10.9 GB | 1.25 GB | 12.8 GB |
1.6 GB free
|
16 | llama.cpp | Run |
| Mistral Small 3.2 24B | 23.6B | Q3_K_M | 10.7 GB | 1.25 GB | 12.6 GB |
1.8 GB free
|
16 | llama.cpp | Run |
Won't fit in VRAM
35 models · some are possible with CPU offload| Model | Params | Quant | Weights | +KV 8K | Total | Over budget | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| Llama 3.3 70B Instruct | 70.6B | Q3_K_M | 32.1 GB | 2.50 GB | 35.2 GB | +20.8 GB | ~1.6 | llama.cpp | Why |
| DeepSeek-R1-Distill-Llama 70B | 70.6B | Q3_K_M | 32.1 GB | 2.50 GB | 35.2 GB | +20.8 GB | ~1.6 | llama.cpp | Why |
| Qwen3.6 35B-A3B | 35.9B MoE / 3.3B act | Q3_K_M | 16.3 GB | 0.16 GB | 17.1 GB | +2.7 GB | ~27 | llama.cpp | Why |
| Ornith 1.5 35B-A3B | 35.9B MoE / 3.3B act | Q3_K_M | 16.3 GB | 0.16 GB | 17.1 GB | +2.7 GB | ~27 | llama.cpp | Why |
| LLM-jp 4 33B Thinking | 33.2B | Q3_K_M | 15.1 GB | 2.00 GB | 17.7 GB | +3.3 GB | ~6.3 | llama.cpp | Why |
| Qwen3 32B | 32.8B | Q3_K_M | 14.9 GB | 2.00 GB | 17.5 GB | +3.1 GB | ~6.5 | llama.cpp | Why |
| DeepSeek-R1-Distill-Qwen 32B | 32.8B | Q3_K_M | 14.9 GB | 2.00 GB | 17.5 GB | +3.1 GB | ~6.5 | llama.cpp | Why |
| Qwen2.5-Coder 32B | 32.8B | Q3_K_M | 14.9 GB | 2.00 GB | 17.5 GB | +3.1 GB | ~6.5 | llama.cpp | Why |
| Olmo 3.1 32B Instruct | 32.2B | Q3_K_M | 14.7 GB | 1.25 GB | 16.5 GB | +2.1 GB | ~7.7 | llama.cpp | Why |
| Nemotron 3.5 Lightning 30B-A3B | 31.6B MoE / 3.2B act | Q3_K_M | 14.4 GB | 0.05 GB | 15.0 GB | +0.6 GB | ~39 | llama.cpp | Why |
| Gemma 4 31B | 31.3B | Q3_K_M | 14.2 GB | 2.03 GB | 16.9 GB | +2.5 GB | ~7.4 | llama.cpp | Why |
| GLM-4.7-Flash 30B-A3B | 31.2B MoE / 3B act | Q3_K_M | 14.2 GB | 0.41 GB | 15.2 GB | +0.8 GB | ~40 | llama.cpp | Why |
| Qwen3 30B-A3B | 30.5B MoE / 3.3B act | Q3_K_M | 13.9 GB | 0.75 GB | 15.2 GB | +0.8 GB | ~36 | llama.cpp | Why |
| Qwen3 Coder 30B-A3B | 30.5B MoE / 3.3B act | Q3_K_M | 13.9 GB | 0.75 GB | 15.2 GB | +0.8 GB | ~36 | llama.cpp | Why |
| Granite 4.1 30B | 28.9B | Q3_K_M | 13.2 GB | 2.00 GB | 15.8 GB | +1.4 GB | ~9.5 | llama.cpp | Why |
| Kimi K3 2.8T-A104B | 2780B MoE / 104B act | Q3_K_M | 1265.4 GB | 0.21 GB | 1266.2 GB | +1251.8 GB | ~0.3 | llama.cpp | Why |
| Qwen3.8 2.4T-A95B | 2446B MoE / 95B act | Q3_K_M | 1113.4 GB | 0.72 GB | 1114.7 GB | +1100.3 GB | ~0.3 | llama.cpp | Why |
| DeepSeek V4 Pro 1.6T-A49B | 1650B MoE / 49B act | Q3_K_M | 751.1 GB | 0.54 GB | 752.2 GB | +737.8 GB | ~0.6 | llama.cpp | Why |
| Kimi K2.6 1T-A32B | 1027B MoE / 32B act | Q3_K_M | 467.5 GB | 0.54 GB | 468.6 GB | +454.2 GB | ~1.0 | llama.cpp | Why |
| GLM-5.2 744B-A40B | 753B MoE / 40B act | Q3_K_M | 342.8 GB | 0.69 GB | 344.0 GB | +329.6 GB | ~0.8 | llama.cpp | Why |
| GLM-5.3 744B-A40B | 753B MoE / 40B act | Q3_K_M | 342.8 GB | 0.69 GB | 344.0 GB | +329.6 GB | ~0.8 | llama.cpp | Why |
| DeepSeek-R1 671B | 671B MoE / 37B act | Q3_K_M | 305.4 GB | 0.54 GB | 306.6 GB | +292.2 GB | ~0.9 | llama.cpp | Why |
| DeepSeek V4.1 Flash 552B-A16B New | 552B MoE / 16B act | Q3_K_M | 251.3 GB | 0.04 GB | 251.9 GB | +237.5 GB | ~2.0 | — | Why |
| Qwen3.5 397B-A17B | 403B MoE / 17B act | Q3_K_M | 183.4 GB | 0.23 GB | 184.3 GB | +169.9 GB | ~1.9 | llama.cpp | Why |
| Ornith 1.5 397B-A17B | 397B MoE / 17B act | Q3_K_M | 180.7 GB | 0.23 GB | 181.5 GB | +167.1 GB | ~1.9 | llama.cpp | Why |
| GLM-5.3-Flash 320B-A18B | 321B MoE / 18B act | Q3_K_M | 146.1 GB | 0.09 GB | 146.8 GB | +132.4 GB | ~1.8 | llama.cpp | Why |
| DeepSeek V4 Flash 284B-A13B | 284B MoE / 13B act | Q3_K_M | 129.3 GB | 0.38 GB | 130.3 GB | +115.9 GB | ~2.6 | llama.cpp | Why |
| Qwen3 235B-A22B | 235B MoE / 22B act | Q3_K_M | 107.0 GB | 1.47 GB | 109.0 GB | +94.6 GB | ~1.5 | llama.cpp | Why |
| Qwen3.8-Flash-Next 180B-A6B | 180B MoE / 6B act | Q3_K_M | 81.9 GB | 0.19 GB | 82.7 GB | +68.3 GB | ~5.9 | llama.cpp | Why |
| Qwen3.5 122B-A10B | 125B MoE / 10B act | Q3_K_M | 56.9 GB | 0.19 GB | 57.7 GB | +43.3 GB | ~3.8 | llama.cpp | Why |
| Nemotron 3 Super 120B-A12B | 124B MoE / 12B act | Q3_K_M | 56.4 GB | 0.06 GB | 57.1 GB | +42.7 GB | ~3.2 | llama.cpp | Why |
| Ling 3.0 Flash 124B-A5B | 124B MoE / 5.1B act | Q3_K_M | 56.4 GB | 0.06 GB | 57.1 GB | +42.7 GB | ~7.5 | llama.cpp | Why |
| Mistral Small 4 119B-A6B | 119B MoE / 6.5B act | Q3_K_M | 54.2 GB | 0.18 GB | 54.9 GB | +40.5 GB | ~5.9 | llama.cpp | Why |
| gpt-oss 120B | 117B MoE / 5.1B act | MXFP4 | 60.5 GB | 0.29 GB | 61.4 GB | +47.0 GB | ~6.4 | llama.cpp | Why |
| Llama 4 Scout 109B-A17B | 109B MoE / 17B act | Q3_K_M | 49.6 GB | 1.50 GB | 51.7 GB | +37.3 GB | ~2.2 | llama.cpp | Why |
Totals = quantised weights + f16 KV cache at 8K tokens + 0.6 GB runtime overhead, against 14.4 GB usable VRAM (16 GB less display and driver reserve). Tokens per second are estimated from 288 GB/s peak memory bandwidth at batch 1 — not measured. Read the method.
Same memory, different speed
10 of the 19 devices have the same 14.4 GB usable, so they run exactly the models above. Generation speed scales with memory bandwidth.
| Device | Bandwidth | Speed vs yours |
|---|---|---|
| GeForce RTX 5080 Blackwell · 2025 |
960 GB/s | 3.33× |
| GeForce RTX 5070 Ti Blackwell · 2025 |
896 GB/s | 3.11× |
| CMP 100-210 Volta · 2021 |
829 GB/s | 2.88× |
| GeForce RTX 4080 Super Ada Lovelace · 2024 |
736 GB/s | 2.56× |
| Tesla P100 16 GB Pascal · 2016 |
732 GB/s | 2.54× |
| GeForce RTX 4080 Ada Lovelace · 2022 |
717 GB/s | 2.49× |
| GeForce RTX 4070 Ti Super Ada Lovelace · 2024 |
672 GB/s | 2.33× |
| Radeon RX 9070 XT RDNA 4 · 2025 |
645 GB/s | 2.24× |
| Radeon RX 9070 RDNA 4 · 2025 |
645 GB/s | 2.24× |
| Radeon RX 7800 XT RDNA 3 · 2023 |
624 GB/s | 2.17× |
What more memory unlocks
At Q4_K_M and 8K context this device fits 38 of 80 models. Each step up, and the largest models it adds:
| Step up | Models fit | Adds, largest first |
|---|---|---|
| 24 GB card e.g. GeForce RTX 4090 |
58 (+20) | Qwen3.6 35B-A3B, Ornith 1.5 35B-A3B, LLM-jp 4 33B Thinking |
| 32 GB card e.g. GeForce RTX 5090 |
58 (+20) | Qwen3.6 35B-A3B, Ornith 1.5 35B-A3B, LLM-jp 4 33B Thinking |
| 48 GB card e.g. RTX 6000 Ada |
60 (+22) | Llama 3.3 70B Instruct, DeepSeek-R1-Distill-Llama 70B, Qwen3.6 35B-A3B |