Your rig
Radeon RX 6700 XT · 12 GB
64 GB system RAM · RDNA 2 · 384 GB/s · 10.6 GB usable VRAM
Which one should I actually run?
Best models for your Radeon RX 6700 XT
What do you want to do? Each row is the best-ranked model on that use case's shortlist that fits at Q2_K and 8K. The ordering is editorial; the sizes and speeds are computed. How the shortlists work.
All
General assistantCodingReasoning & mathsWritingVisionAgents & tool useTranslationLow latencyLong context
| Recommendation | Model | Why | Quant · total | Tok/s est. | |
|---|---|---|---|---|---|
| Best overall | Gemma 4 12B Google · Apache 2.0 |
The 12 GB generalist: text, image and audio in, 140+ languages. | Q2_K · 6.1 GB | 50 | Runs great |
| Best for coding | Qwen3.5 9B Alibaba · Apache 2.0 |
The best coding model for 8–12 GB cards. | Q2_K · 4.6 GB | 62 | Runs great |
| Best reasoning | Qwen3.5 9B Alibaba · Apache 2.0 |
Thinking mode on a 12 GB card — ahead of the 2025 R1 distills. | Q2_K · 4.6 GB | 62 | Runs great |
| Best for writing | Gemma 4 12B Google · Apache 2.0 |
The 12 GB writing pick; 140+ languages. | Q2_K · 6.1 GB | 50 | Runs great |
| Best vision | Gemma 4 12B Google · Apache 2.0 |
Image and audio understanding on 12 GB. | Q2_K · 6.1 GB | 50 | Runs great |
| Best for agents | Granite 4.1 8B IBM · Apache 2.0 |
Enterprise-grade tool calling in a dense 8B, no thinking overhead. | Q2_K · 5.3 GB | 68 | Runs great |
| Best translation | Gemma 4 12B Google · Apache 2.0 |
Broad language coverage on a 12 GB card. | Q2_K · 6.1 GB | 50 | Runs great |
| Fastest good model | LFM2.5 8B-A1B Liquid AI · LFM Open License v1.0 |
1.5B active; designed for laptops without a GPU. | Q2_K · 4.0 GB | 153 | Runs great |
| Best long context | Qwen3.5 9B Alibaba · Apache 2.0 |
262K native in 8–12 GB. | Q2_K · 8.4 GB at 128K context |
62 | Runs great |
Everything that fits
Runs great
37 of 76 · fits with headroom for a longer prompt| Model | Params | Quant | Weights | +KV 8K | Total | VRAM headroom | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| Qwen3 14B | 14.8B | Q2_K | 5.8 GB | 1.25 GB | 7.6 GB |
3.0 GB free
|
40 | Ollama | Run |
| DeepSeek-R1-Distill-Qwen 14B | 14.8B | Q2_K | 5.8 GB | 1.50 GB | 7.9 GB |
2.7 GB free
|
40 | Ollama | Run |
| Qwen2.5-Coder 14B | 14.8B | Q2_K | 5.8 GB | 1.50 GB | 7.9 GB |
2.7 GB free
|
40 | Ollama | Run |
| Phi-4 14B | 14.7B | Q2_K | 5.7 GB | 1.56 GB | 7.9 GB |
2.7 GB free
|
41 | Ollama | Run |
| Ministral 3 14B | 13.9B | Q2_K | 5.4 GB | 1.25 GB | 7.3 GB |
3.3 GB free
|
43 | Ollama | Run |
| Gemma 3 12B | 12.2B | Q2_K | 4.8 GB | 0.81 GB | 6.2 GB |
4.4 GB free
|
49 | Ollama | Run |
| Mistral NeMo 12B | 12.2B | Q2_K | 4.8 GB | 1.25 GB | 6.6 GB |
4.0 GB free
|
49 | Ollama | Run |
| Gemma 4 12B | 12B | Q2_K | 4.7 GB | 0.81 GB | 6.1 GB |
4.5 GB free
|
50 | Ollama | Run |
| Qwen3.5 9B | 9.65B | Q2_K | 3.8 GB | 0.25 GB | 4.6 GB |
6.0 GB free
|
62 | Ollama | Run |
| Ornith 1.5 9B New this week | 9.41B | Q2_K | 3.7 GB | 0.25 GB | 4.5 GB |
6.1 GB free
|
63 | Ollama | Run |
| Ministral 3 8B | 8.92B | Q2_K | 3.5 GB | 1.06 GB | 5.1 GB |
5.5 GB free
|
67 | Ollama | Run |
| Granite 4.1 8B | 8.79B | Q2_K | 3.4 GB | 1.25 GB | 5.3 GB |
5.3 GB free
|
68 | Ollama | Run |
| LFM2.5 8B-A1B | 8.47B MoE / 1.5B act | Q2_K | 3.3 GB | 0.09 GB | 4.0 GB |
6.6 GB free
|
153 | Ollama | Run |
| Fara 7B | 8.29B | Q2_K | 3.2 GB | 0.44 GB | 4.3 GB |
6.3 GB free
|
72 | llama.cpp | Run |
| Qwen3 8B | 8.19B | Q2_K | 3.2 GB | 1.13 GB | 4.9 GB |
5.7 GB free
|
73 | Ollama | Run |
| Llama 3.1 8B Instruct | 8.03B | Q2_K | 3.1 GB | 1.00 GB | 4.7 GB |
5.9 GB free
|
74 | Ollama | Run |
| Gemma 4 E4B | 8.0B | Q2_K | 3.1 GB | 0.14 GB | 3.9 GB |
6.7 GB free
|
75 | Ollama | Run |
| Ling 3.0 Tiny 7.9B-A1.3B New | 7.9B MoE / 1.3B act | Q2_K | 3.1 GB | 0.05 GB | 3.7 GB |
6.9 GB free
|
176 | llama.cpp | Run |
| DeepSeek-R1-Distill-Qwen 7B | 7.62B | Q2_K | 3.0 GB | 0.44 GB | 4.0 GB |
6.6 GB free
|
78 | Ollama | Run |
| Qwen2.5-Coder 7B | 7.62B | Q2_K | 3.0 GB | 0.44 GB | 4.0 GB |
6.6 GB free
|
78 | Ollama | Run |
| Olmo 3 7B Instruct | 7.3B | Q2_K | 2.8 GB | 2.50 GB | 5.9 GB |
4.7 GB free
|
82 | Ollama | Run |
| Mistral 7B Instruct v0.3 | 7.25B | Q2_K | 2.8 GB | 1.00 GB | 4.4 GB |
6.2 GB free
|
82 | Ollama | Run |
| Gemma 4 E2B | 5.1B | Q2_K | 2.0 GB | 0.07 GB | 2.7 GB |
7.9 GB free
|
117 | Ollama | Run |
| Qwen3.5 4B | 4.66B | Q2_K | 1.8 GB | 0.25 GB | 2.7 GB |
7.9 GB free
|
128 | Ollama | Run |
| Gemma 3 4B | 4.3B | Q2_K | 1.7 GB | 0.30 GB | 2.6 GB |
8.0 GB free
|
139 | Ollama | Run |
| Qwen3 4B | 4.02B | Q2_K | 1.6 GB | 1.13 GB | 3.3 GB |
7.3 GB free
|
148 | Ollama | Run |
| Ministral 3 3B | 3.85B | Q2_K | 1.5 GB | 0.81 GB | 2.9 GB |
7.7 GB free
|
155 | Ollama | Run |
| Phi-4-mini 3.8B | 3.84B | Q2_K | 1.5 GB | 1.00 GB | 3.1 GB |
7.5 GB free
|
155 | Ollama | Run |
| Granite 4.1 3B | 3.4B | Q2_K | 1.3 GB | 0.63 GB | 2.6 GB |
8.0 GB free
|
175 | Ollama | Run |
| Llama 3.2 3B Instruct | 3.21B | Q2_K | 1.3 GB | 0.88 GB | 2.7 GB |
7.9 GB free
|
186 | Ollama | Run |
| LFM2.5 2.6B New | 2.7B | Q2_K | 1.1 GB | 0.13 GB | 1.8 GB |
8.8 GB free
|
221 | llama.cpp | Run |
| Qwen3.5 2B | 2.27B | Q2_K | 0.9 GB | 0.09 GB | 1.6 GB |
9.0 GB free
|
263 | Ollama | Run |
| Qwen3 1.7B | 1.72B | Q2_K | 0.7 GB | 0.88 GB | 2.1 GB |
8.5 GB free
|
347 | Ollama | Run |
| Llama 3.2 1B Instruct | 1.24B | Q2_K | 0.5 GB | 0.25 GB | 1.3 GB |
9.3 GB free
|
481 | Ollama | Run |
| Gemma 3 1B | 1.0B | Q2_K | 0.4 GB | 0.04 GB | 1.0 GB |
9.6 GB free
|
596 | Ollama | Run |
| Qwen3.5 0.8B | 0.87B | Q2_K | 0.3 GB | 0.09 GB | 1.0 GB |
9.6 GB free
|
685 | Ollama | Run |
| Qwen3 0.6B | 0.6B | Q2_K | 0.2 GB | 0.88 GB | 1.7 GB |
8.9 GB free
|
993 | Ollama | Run |
Won't fit in VRAM
39 models · some are possible with CPU offload| Model | Params | Quant | Weights | +KV 8K | Total | Over budget | Tok/s | Runtime | |
|---|---|---|---|---|---|---|---|---|---|
| Qwen3.5 122B-A10B | 125B MoE / 10B act | Q2_K | 48.7 GB | 0.19 GB | 49.5 GB | +38.9 GB | ~4.3 | llama.cpp | Why |
| Nemotron 3 Super 120B-A12B | 124B MoE / 12B act | Q2_K | 48.4 GB | 0.06 GB | 49.0 GB | +38.4 GB | ~3.6 | llama.cpp | Why |
| Ling 3.0 Flash 124B-A5B New | 124B MoE / 5.1B act | Q2_K | 48.4 GB | 0.06 GB | 49.0 GB | +38.4 GB | ~8.5 | llama.cpp | Why |
| Mistral Small 4 119B-A6B | 119B MoE / 6.5B act | Q2_K | 46.4 GB | 0.18 GB | 47.2 GB | +36.6 GB | ~6.7 | llama.cpp | Why |
| gpt-oss 120B | 117B MoE / 5.1B act | MXFP4 | 60.5 GB | 0.29 GB | 61.4 GB | +50.8 GB | ~6.1 | llama.cpp | Why |
| Llama 4 Scout 109B-A17B | 109B MoE / 17B act | Q2_K | 42.5 GB | 1.50 GB | 44.6 GB | +34.0 GB | ~2.5 | llama.cpp | Why |
| Llama 3.3 70B Instruct | 70.6B | Q2_K | 27.5 GB | 2.50 GB | 30.6 GB | +20.0 GB | ~1.7 | llama.cpp | Why |
| DeepSeek-R1-Distill-Llama 70B | 70.6B | Q2_K | 27.5 GB | 2.50 GB | 30.6 GB | +20.0 GB | ~1.7 | llama.cpp | Why |
| Qwen3.6 35B-A3B | 35.9B MoE / 3.3B act | Q2_K | 14.0 GB | 0.16 GB | 14.8 GB | +4.2 GB | ~27 | llama.cpp | Why |
| Ornith 1.5 35B-A3B New this week | 35.9B MoE / 3.3B act | Q2_K | 14.0 GB | 0.16 GB | 14.8 GB | +4.2 GB | ~27 | llama.cpp | Why |
| LLM-jp 4 33B Thinking New this week | 33.2B | Q2_K | 12.9 GB | 2.00 GB | 15.5 GB | +4.9 GB | ~5.9 | llama.cpp | Why |
| Qwen3 32B | 32.8B | Q2_K | 12.8 GB | 2.00 GB | 15.4 GB | +4.8 GB | ~6.0 | llama.cpp | Why |
| DeepSeek-R1-Distill-Qwen 32B | 32.8B | Q2_K | 12.8 GB | 2.00 GB | 15.4 GB | +4.8 GB | ~6.0 | llama.cpp | Why |
| Qwen2.5-Coder 32B | 32.8B | Q2_K | 12.8 GB | 2.00 GB | 15.4 GB | +4.8 GB | ~6.0 | llama.cpp | Why |
| Olmo 3.1 32B Instruct | 32.2B | Q2_K | 12.6 GB | 1.25 GB | 14.4 GB | +3.8 GB | ~7.0 | llama.cpp | Why |
| Nemotron 3.5 Lightning 30B-A3B New | 31.6B MoE / 3.2B act | Q2_K | 12.3 GB | 0.05 GB | 13.0 GB | +2.4 GB | ~35 | llama.cpp | Why |
| Gemma 4 31B | 31.3B | Q2_K | 12.2 GB | 2.03 GB | 14.8 GB | +4.2 GB | ~6.6 | llama.cpp | Why |
| GLM-4.7-Flash 30B-A3B | 31.2B MoE / 3B act | Q2_K | 12.2 GB | 0.41 GB | 13.2 GB | +2.6 GB | ~36 | llama.cpp | Why |
| Qwen3 30B-A3B | 30.5B MoE / 3.3B act | Q2_K | 11.9 GB | 0.75 GB | 13.2 GB | +2.6 GB | ~32 | llama.cpp | Why |
| Qwen3 Coder 30B-A3B | 30.5B MoE / 3.3B act | Q2_K | 11.9 GB | 0.75 GB | 13.2 GB | +2.6 GB | ~32 | llama.cpp | Why |
| Muse Glimmer 30B New | 29.8B | Q2_K | 11.6 GB | 0.18 GB | 12.4 GB | +1.8 GB | ~11 | llama.cpp | Why |
| Granite 4.1 30B | 28.9B | Q2_K | 11.3 GB | 2.00 GB | 13.9 GB | +3.3 GB | ~8.0 | llama.cpp | Why |
| Qwen3.8 27B New this week | 27.8B | Q2_K | 10.8 GB | 0.50 GB | 11.9 GB | +1.3 GB | ~13 | llama.cpp | Why |
| Qwen3.6 27B | 27.8B | Q2_K | 10.8 GB | 0.50 GB | 11.9 GB | +1.3 GB | ~13 | llama.cpp | Why |
| Gemma 3 27B | 27.4B | Q2_K | 10.7 GB | 1.03 GB | 12.3 GB | +1.7 GB | ~12 | llama.cpp | Why |
| Gemma 4 26B-A4B | 26.5B MoE / 3.8B act | Q2_K | 10.3 GB | 0.51 GB | 11.4 GB | +0.8 GB | ~42 | llama.cpp | Why |
| Devstral Small 2 24B | 24B | Q2_K | 9.4 GB | 1.25 GB | 11.2 GB | +0.6 GB | ~18 | llama.cpp | Why |
| Mistral Small 3.2 24B | 23.6B | Q2_K | 9.2 GB | 1.25 GB | 11.1 GB | +0.5 GB | ~20 | llama.cpp | Why |
| gpt-oss 20B | 20.9B MoE / 3.6B act | MXFP4 | 10.8 GB | 0.19 GB | 11.6 GB | +1.0 GB | ~32 | llama.cpp | Why |
| Kimi K3 2.8T-A104B New | 2780B MoE / 104B act | Q2_K | 1084.2 GB | 0.21 GB | 1085.0 GB | +1074.4 GB | ~0.3 | llama.cpp | Why |
| Qwen3.8 2.4T-A95B New | 2446B MoE / 95B act | Q2_K | 953.9 GB | 0.72 GB | 955.2 GB | +944.6 GB | ~0.4 | llama.cpp | Why |
| DeepSeek V4 Pro 1.6T-A49B New | 1650B MoE / 49B act | Q2_K | 643.5 GB | 0.54 GB | 644.6 GB | +634.0 GB | ~0.7 | llama.cpp | Why |
| Kimi K2.6 1T-A32B | 1027B MoE / 32B act | Q2_K | 400.5 GB | 0.54 GB | 401.7 GB | +391.1 GB | ~1.1 | llama.cpp | Why |
| GLM-5.2 744B-A40B | 753B MoE / 40B act | Q2_K | 293.7 GB | 0.69 GB | 294.9 GB | +284.3 GB | ~0.9 | llama.cpp | Why |
| DeepSeek-R1 671B | 671B MoE / 37B act | Q2_K | 261.7 GB | 0.54 GB | 262.8 GB | +252.2 GB | ~1.0 | llama.cpp | Why |
| Qwen3.5 397B-A17B | 403B MoE / 17B act | Q2_K | 157.2 GB | 0.23 GB | 158.0 GB | +147.4 GB | ~2.2 | llama.cpp | Why |
| Ornith 1.5 397B-A17B New this week | 397B MoE / 17B act | Q2_K | 154.8 GB | 0.23 GB | 155.7 GB | +145.1 GB | ~2.2 | llama.cpp | Why |
| DeepSeek V4 Flash 284B-A13B New | 284B MoE / 13B act | Q2_K | 110.8 GB | 0.38 GB | 111.7 GB | +101.1 GB | ~3.0 | llama.cpp | Why |
| Qwen3 235B-A22B | 235B MoE / 22B act | Q2_K | 91.6 GB | 1.47 GB | 93.7 GB | +83.1 GB | ~1.8 | llama.cpp | Why |
Totals = quantised weights + f16 KV cache at 8K tokens + 0.6 GB runtime overhead, against 10.6 GB usable VRAM (12 GB less display and driver reserve). Tokens per second are estimated from 384 GB/s peak memory bandwidth at batch 1 — not measured. Read the method.