Lenovo ThinkStation PGX GB10 128GB
128 GB unified · 273 GB/s
In short
The Lenovo ThinkStation PGX GB10 128GB has 128 GB of unified memory, of which the GPU can use 119 GB by default, with 273 GB/s of memory bandwidth. At 128k context it runs 80 of the 129 open models we track fully in memory; the most capable is DeepSeek V4 Flash at UD-Q2_K_XL, at about 6 tok/s (estimated). We have no current price for the Lenovo ThinkStation PGX GB10 128GB in the UK.
Updated . Speeds are estimates; how they are worked out.
Prices
Every retailer we track in the UK, cheapest first. Buy links may earn us a commission.
No retailer in this market lists it right now.
What it runs
Every model in the catalogue, with the best quant for this hardware. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @128k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | fits | 102.4 GB | 312k557k with q8 cache | 6tok/sest. | 68 tok/s | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | fits | 104.2 GB | 1018k1024k with q8 cache | 10tok/sest. | 114 tok/s | – |
| Qwen 3.8 Flash Next Alibaba Qwen | Q4_K_M | fits | 115.9 GB | 232k256k with q8 cache | 2tok/sest. | 196 tok/s | – |
| Qwen 3.8 27B Alibaba Qwen | Q8_0 | fits | 36.6 GB | 256k | 5tok/sest. | 343 tok/s | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | fits | 102.6 GB | 194k200k with q8 cache | 3tok/sest. | 229 tok/s | chat |
| Qwen 3.5 122B-A10B Alibaba Qwen | Q6_K | fits | 98.7 GB | 256k | 9tok/sest. | 610 tok/s | – |
| Qwen 3.6 27B Alibaba Qwen | Q8_0 | fits | 36.2 GB | 256k | 5tok/sest. | 343 tok/s | – |
| Qwen3 Coder Next Alibaba Qwen | Q8_0 | fits | 83.5 GB | 256k | 14tok/sest. | 1352 tok/s | chat, coding |
| gpt-oss 120B OpenAI | MXFP4 | fits | 65.0 GB | 128k | 13tok/sest. | 904 tok/s | chat, coding |
| Nemotron 3 Super 120B-A12B NVIDIA | UD-Q6_K | fits | 109.5 GB | 256k | 8tok/sest. | 856 tok/s | – |
| Gemma 4 31B | Q8_0 | fits | 42.6 GB | 256k | 4tok/sest. | 326 tok/s | – |
| Mistral Small 4 119B Mistral AI | UD-Q6_K | fits | 96.9 GB | 923k1024k with q8 cache | 11tok/sest. | 529 tok/s | chat |
| Qwen 3.6 35B-A3B Alibaba Qwen | Q8_0 | fits | 38.4 GB | 256k | 17tok/sest. | 1667 tok/s | chat, coding, long documents |
| Qwen 3.5 27B Alibaba Qwen | Q8_0 | fits | 36.2 GB | 256k | 5tok/sest. | 343 tok/s | – |
| Gemma 4 26B-A4B | Q8_0 | fits | 29.2 GB | 256k | 14tok/sest. | 2024 tok/s | chat, coding |
| GLM 4.5 Air Z.ai | Q6_K | fits | 116.7 GB | 128k | 3tok/sest. | 159 tok/s | – |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q8_0 | fits | 38.4 GB | 256k | 17tok/sest. | 1667 tok/s | chat, coding, long documents |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q8_0 | fits | 83.5 GB | 256k | 16tok/sest. | 1477 tok/s | chat, coding, long documents |
| GLM 4.7 Flash Z.ai | Q8_0 | fits | 37.7 GB | 198k | 8tok/sest. | 786 tok/s | chat |
| Devstral Small 2 24B Mistral AI | Q8_0 | fits | 44.8 GB | 384k | 4tok/sest. | 307 tok/s | – |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | UD-Q6_K_XL | fits | 34.8 GB | 256k | 24tok/sest. | 2242 tok/s | chat, coding, long documents |
| Kimi Linear 48B-A3B Moonshot AI | Q8_0 | fits | 51.1 GB | 1024k | 24tok/sest. | 1251 tok/s | chat, coding, long documents |
| gpt-oss 20B OpenAI | MXFP4 | fits | 15.7 GB | 128k | 18tok/sest. | 1310 tok/s | chat, coding, agents, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q8_0 | fits | 43.7 GB | 256k | 6tok/sest. | 485 tok/s | chat |
| Qwen3 32B Alibaba Qwen | Q8_0 | fits | 65.9 GB | 40k | 2tok/sest. | 138 tok/s | – |
| Gemma 4 12B | Q8_0 | fits | 15.6 GB | 256k | 11tok/sest. | 841 tok/s | – |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 14.4 GB | 256k | 12tok/sest. | 1000 tok/s | chat |
| Gemma 3 27B | Q8_0 | fits | 38.6 GB | 128k | 4tok/sest. | 420 tok/s | – |
| Mistral Small 3.2 24B Mistral AI | Q8_0 | fits | 44.8 GB | 128k | 4tok/sest. | 307 tok/s | – |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q8_0 | fits | 65.9 GB | 128k | 2tok/sest. | 184 tok/s | – |
| Llama 3.3 70B Meta | Q8_0 | fits | 111.3 GB | 128k | 1tok/sest. | 89 tok/s | – |
| Ministral 3 14B Mistral AI | Q8_0 | fits | 34.8 GB | 256k | 5tok/sest. | 394 tok/s | – |
| Qwen3 14B Alibaba Qwen | Q8_0 | fits | 36.1 GB | 40k | 5tok/sest. | 335 tok/s | – |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q8_0 | fits | 27.6 GB | 128k | 6tok/sest. | 507 tok/s | chat |
| Qwen3 8B Alibaba Qwen | Q8_0 | fits | 27.6 GB | 40k | 6tok/sest. | 507 tok/s | chat |
| Phi-4 14B Microsoft | Q8_0 | fits | 41.0 GB | 16k | 4tok/sest. | 336 tok/s | – |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 9.7 GB | 256k | 19tok/sest. | 1558 tok/s | chat, coding, long documents |
| Ministral 3 8B Mistral AI | Q8_0 | fits | 26.9 GB | 256k | 6tok/sest. | 513 tok/s | chat |
| Granite 4.2 8B IBM | Q8_0 | fits | 30.2 GB | 128k | 6tok/sest. | 461 tok/s | chat |
| Gemma 4 E4B | Q8_0 | fits | 12.6 GB | 128k | 14tok/sest. | 1402 tok/s | chat |
| LFM2.5 8B-A1B Liquid AI | Q8_0 | fits | 15.8 GB | 125k | 14tok/sest. | 1874 tok/s | chat, coding |
| Llama 3.1 8B Meta | Q6_K | fits | 23.6 GB | 128k | 7tok/sest. | 553 tok/s | chat |
| Qwen2.5 0.5B Alibaba Qwen | Q8_0 | fits | 3.4 GB | 32k | 79tok/sest. | 4211 tok/s | chat, coding, long documents |
| Qwen3.5 0.8B Alibaba Qwen | Q8_0 | fits | 3.7 GB | 256k | 69tok/sest. | 5616 tok/s | chat, coding, agents, long documents |
| Qwen3.5 2B Alibaba Qwen | Q5_K_M | fits | 4.3 GB | 256k | 55tok/sest. | 3591 tok/s | chat, coding, agents, long documents |
| LFM3 1.7B Liquid AI | Q8_0 | fits | 4.7 GB | 256k | 49tok/sest. | 4154 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 1.5B DeepSeek | Q8_0 | fits | 6.7 GB | 128k | 30tok/sest. | 1882 tok/s | chat, coding, long documents |
| Mellum2.1 12B A2.5B Thinking JetBrains | Q8_0 | fits | 15.3 GB | 128k | 22tok/sest. | 2168 tok/s | chat, coding, agents, long documents |
| Spark X2.5 4B XHToken | mlx-4bit | fits | 8.8 GB | 1024k | 21tok/sest. | 1540 tok/s | chat, coding, agents, long documents |
| MiniCPM5 2B openbmb | Q8_0 | fits | 9.2 GB | 128k | 20tok/sest. | 1012 tok/s | chat, coding, agents, long documents |
| FrogNano 4B 2609 Microsoft | Q8_0 | fits | 9.8 GB | 256k | 19tok/sest. | 1558 tok/s | chat, coding, long documents |
| LFM2.5 2.6B Liquid AI | Q8_0 | fits | 11.6 GB | 128k | 16tok/sest. | 1298 tok/s | chat, coding, agents, long documents |
| LFM2.5 VL 3B Liquid AI | Q8_0 | fits | 11.6 GB | 32k | 16tok/sest. | 1248 tok/s | chat, coding, agents, long documents |
| Kimi VL A3B Thinking 2506 Moonshot AI | Q8_0 | fits | 21.1 GB | 128k | 14tok/sest. | 1361 tok/s | chat, coding |
| Kimi VL A3B Moonshot AI | mlx-8bit | fits | 21.8 GB | 128k | 14tok/sest. | 1006 tok/s | chat, coding |
| GLM 4 9B 0414 Z.ai | mlx-5bit | fits | 14.1 GB | 32k | 12tok/sest. | 452 tok/s | chat |
| granite vision 4.1 4b IBM | Q8_0 | fits | 14.8 GB | 128k | 12tok/sest. | 801 tok/s | chat, coding |
| granite 4.2 3b IBM | Q8_0 | fits | 15.1 GB | 128k | 12tok/sest. | 817 tok/s | chat, coding |
| GLM 4.6V Flash Z.ai | Q8_0 | fits | 15.8 GB | 128k | 11tok/sest. | 439 tok/s | chat |
| DeepSeek R1 Distill Qwen 7B DeepSeek | Q8_0 | fits | 16.0 GB | 128k | 11tok/sest. | 672 tok/s | chat |
| GLM 4.1V 9B Thinking Z.ai | mlx-8bit | fits | 17.4 GB | 64k | 10tok/sest. | 439 tok/s | – |
| Ministral 3 3B Instruct 2512 Mistral AI | Q8_0 | fits | 17.9 GB | 256k | 10tok/sest. | 783 tok/s | chat, coding |
| LFM2.5 24B A2B Liquid AI | Q8_0 | fits | 35.2 GB | 128k | 8tok/sest. | 1095 tok/s | chat, coding |
| DeepSeek R1 Distill Llama 8B DeepSeek | Q8_0 | fits | 25.4 GB | 128k | 7tok/sest. | 553 tok/s | chat |
| Ministral 3 8B Reasoning 2512 Mistral AI | Q8_0 | fits | 26.9 GB | 256k | 6tok/sest. | 513 tok/s | chat |
| Qwen3.8 27B OBLITERATED OBLITERATUS | mlx-4bit | fits | 28.8 GB | 256k | 6tok/sest. | 343 tok/s | – |
| Muse Glimmer 30B meta-models | Q8_0 | fits | 33.3 GB | 128k | 5tok/sest. | 373 tok/s | – |
| Ministral 3 14B Reasoning 2512 Mistral AI | Q8_0 | fits | 34.8 GB | 256k | 5tok/sest. | 394 tok/s | – |
| DeepSeek R1 Distill Qwen 14B DeepSeek | Q8_0 | fits | 40.1 GB | 128k | 4tok/sest. | 297 tok/s | – |
| Devstral Small 2507 Mistral AI | Q8_0 | fits | 44.8 GB | 128k | 4tok/sest. | 310 tok/s | – |
| Magistral Small 2506 Mistral AI | Q8_0 | fits | 44.8 GB | 40k | 4tok/sest. | 310 tok/s | – |
| Magistral Small 2507 Mistral AI | Q8_0 | fits | 44.8 GB | 40k | 4tok/sest. | 310 tok/s | – |
| Magistral Small 2509 Mistral AI | Q8_0 | fits | 44.8 GB | 128k | 4tok/sest. | 307 tok/s | – |
| GLM 4.5V Z.ai | Q4_K_M | fits | 83.7 GB | 64k | 3tok/sest. | 158 tok/s | – |
| GLM 4.6V Z.ai | Q4_K_M | fits | 90.1 GB | 128k | 3tok/sest. | 158 tok/s | – |
| granite 4.2 30b IBM | Q8_0 | fits | 62.4 GB | 128k | 3tok/sest. | 219 tok/s | – |
| DeepSeek R1 Distill Llama 70B DeepSeek | Q8_0 | fits | 111.3 GB | 128k | 1tok/sest. | 89 tok/s | – |
| Devstral 2 123B Instruct 2512 Mistral AI | Q4_K_M | fits | 115.2 GB | 138k256k with q8 cache | 1tok/sest. | 52 tok/s | – |
| Kimi Dev 72B Moonshot AI | Q8_0 | fits | 113.4 GB | 128k | 1tok/sest. | 88 tok/s | – |
| Mistral Medium 3.5 128B Mistral AI | Q4_K_M | fits | 115.2 GB | 138k256k with q8 cache | 1tok/sest. | 52 tok/s | – |
| DeepSeek R1 DeepSeek | Q4_K_M | won't run | 386.7 GB | – | – | – | – |
| DeepSeek R1 0528 DeepSeek | Q4_K_M | won't run | 387.2 GB | – | – | – | – |
| DeepSeek R1 Zero DeepSeek | Q4_K_M | won't run | 386.7 GB | – | – | – | – |
| DeepSeek V3 DeepSeek | Q4_K_M | won't run | 386.7 GB | – | – | – | – |
| DeepSeek V3 0324 DeepSeek | Q4_K_M | won't run | 386.7 GB | – | – | – | – |
| DeepSeek V3.1 DeepSeek | Q4_K_M | won't run | 387.6 GB | – | – | – | – |
| DeepSeek V3.1 Terminus DeepSeek | Q4_K_M | won't run | 387.6 GB | – | – | – | – |
| DeepSeek V3.2 DeepSeek | Q4_K_M | won't run | 387.6 GB | – | – | – | – |
| DeepSeek V4 Flash 0731 DeepSeek | MXFP4 | won't run | 157.8 GB | – | – | – | – |
| DeepSeek V4 Flash Vision Exp DeepSeek | MXFP4 | won't run | 166.6 GB | – | – | – | – |
| GLM 4.5 Z.ai | Q4_K_M | won't run | 249.0 GB | – | – | – | – |
| GLM 4.6 Z.ai | Q4_K_M | won't run | 248.2 GB | – | – | – | – |
| GLM 4.7 Z.ai | Q4_K_M | won't run | 249.0 GB | – | – | – | – |
| GLM 5 Z.ai | MXFP4 | won't run | 394.8 GB | – | – | – | – |
| GLM 5.1 Z.ai | MXFP4 | won't run | 432.4 GB | – | – | – | – |
| GLM 5.2 Z.ai | UD-Q4_K_M | won't run | 446.3 GB | – | – | – | – |
| GLM 5.3 Z.ai | Q8_0 | won't run | 758.7 GB | – | – | – | – |
| Kimi K2 Moonshot AI | Q4_K_M | won't run | 588.2 GB | – | – | – | – |
| Kimi K2 Instruct 0905 Moonshot AI | Q4_K_M | won't run | 588.6 GB | – | – | – | – |
| Kimi K2 Thinking Moonshot AI | Q4_K_M | won't run | 588.6 GB | – | – | – | – |
| Kimi K2.5 Moonshot AI | Q4_K_M | won't run | 588.6 GB | – | – | – | – |
| MiniMax M2 MiniMax | Q4_K_M | won't run | 161.3 GB | – | – | – | – |
| MiniMax M2.1 MiniMax | Q4_K_M | won't run | 161.3 GB | – | – | – | – |
| MiniMax M2.5 MiniMax | MXFP4 | won't run | 147.7 GB | 12k22k with q8 cache | – | – | – |
| MiniMax M3 MiniMax | MXFP4 | won't run | 256.1 GB | – | – | – | – |
Specifications
From the manufacturer's published specifications; sources are linked where we have them.
Compare with
Questions
- How much memory does the Lenovo ThinkStation PGX GB10 128GB have for AI models?
- The Lenovo ThinkStation PGX GB10 128GB has 128 GB of unified memory, of which the GPU can use 119 GB by default, with 273 GB/s of memory bandwidth.
- What is the best model the Lenovo ThinkStation PGX GB10 128GB can run locally?
- DeepSeek V4 Flash (UD-Q2_K_XL) fits fully in its memory at 128k context and generates at about 6 tok/s, an estimate from its memory bandwidth. 80 of the 129 models we track fit at that context.
- Can the Lenovo ThinkStation PGX GB10 128GB run DeepSeek R1?
- No: even at Q4_K_M, DeepSeek R1 needs about 387 GB at 128k context, more than the Lenovo ThinkStation PGX GB10 128GB and system RAM can hold.
- How much does the Lenovo ThinkStation PGX GB10 128GB cost in the UK?
- We have no current price for the Lenovo ThinkStation PGX GB10 128GB in the UK.
Speeds on this page are for 128k context. See the best hardware for local AI at every budget.