Intel · Xe2 (Battlemage), BMG-G21 · 24 GB VRAM tier
Best AI Models for Intel Arc Pro B60 (24 GB)
Runs 14B and 35B MoE models well; 27–31B fits but is slow (13 tok/s).
24 GB GDDR6456 GB/s99 TFLOPS FP16200 WXe2 (Battlemage), BMG-G21 2025
Workstation Battlemage card; runs LLMs via IPEX-LLM / llama.cpp (Vulkan/SYCL). 24 GB at a low price, but the software stack is less mature than CUDA/ROCm.
| # | Model | Speed | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 48.3 t/s | 48.3 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 55.4 t/s | 55.4 tok/s | Q4_K_M · 16.5 GB | |
| 3 | Qwen3 30B A3B Thinking 2507Alibaba Q4_K_M · 19.4 GB · 53.7 t/s | 53.7 tok/s | Q4_K_M · 19.4 GB | |
| 4 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 61.9 t/s | 61.9 tok/s | Q4_K_M · 13.1 GB | |
| 5 | Q4_K_M · 19.3 GB · 59.7 t/s | 59.7 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Coding Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 48.3 t/s | 48.3 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 55.4 t/s | 55.4 tok/s | Q4_K_M · 16.5 GB | |
| 3 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 61.9 t/s | 61.9 tok/s | Q4_K_M · 13.1 GB | |
| 4 | Q4_K_M · 7.9 GB · 43.0 t/s | 43.0 tok/s | Q4_K_M · 7.9 GB | |
| 5 | Q4_K_M · 19.3 GB · 59.7 t/s | 59.7 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Reasoning Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 48.3 t/s | 48.3 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 55.4 t/s | 55.4 tok/s | Q4_K_M · 16.5 GB | |
| 3 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 35.8 t/s | 35.8 tok/s | Q4_K_M · 6.6 GB | |
| 4 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 61.9 t/s | 61.9 tok/s | Q4_K_M · 13.1 GB | |
| 5 | Q4_K_M · 19.3 GB · 59.7 t/s | 59.7 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Math Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 48.3 t/s | 48.3 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Qwen3.8 27BAlibaba Q4_K_M · 17.9 GB · 13.1 t/s | 13.1 tok/s | Q4_K_M · 17.9 GB | |
| 3 | Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 13.1 t/s | 13.1 tok/s | Q4_K_M · 17.9 GB | |
| 4 | Q4_K_M · 19.3 GB · 59.7 t/s | 59.7 tok/s | Q4_K_M · 19.3 GB | |
| 5 | GLM 4.7 FlashZ.ai Q4_K_M · 19.9 GB · 45.0 t/s | 45.0 tok/s | Q4_K_M · 19.9 GB |
| # | Model | Speed | Long context Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Q4_K_M · 16.5 GB · 55.4 t/s | 55.4 tok/s | Q4_K_M · 16.5 GB | |
| 2 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 35.8 t/s | 35.8 tok/s | Q4_K_M · 6.6 GB | |
| 3 | Qwen3.5 4BAlibaba Q4_K_M · 3.4 GB · 69.9 t/s | 69.9 tok/s | Q4_K_M · 3.4 GB | |
| 4 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 61.9 t/s | 61.9 tok/s | Q4_K_M · 13.1 GB | |
| 5 | LFM2.5 24B A2BLiquid AI Q4_K_M · 15.0 GB · 68.2 t/s | 68.2 tok/s | Q4_K_M · 15.0 GB |
| # | Model | Speed ↓ | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 69.9 t/s | 69.9 tok/s | Q4_K_M · 3.4 GB | |
| 2 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 61.9 t/s | 61.9 tok/s | Q4_K_M · 13.1 GB | |
| 3 | Q4_K_M · 19.3 GB · 59.7 t/s | 59.7 tok/s | Q4_K_M · 19.3 GB | |
| 4 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 55.4 t/s | 55.4 tok/s | Q4_K_M · 16.5 GB | |
| 5 | Qwen3 30B A3B Thinking 2507Alibaba Q4_K_M · 19.4 GB · 53.7 t/s | 53.7 tok/s | Q4_K_M · 19.4 GB |
| # | Model | Speed | Vision Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.6 35B A3BAlibabaHow to run → Q4_K_M · 22.0 GB · 48.3 t/s | 48.3 tok/s | Q4_K_M · 22.0 GB | |
| 2 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 55.4 t/s | 55.4 tok/s | Q4_K_M · 16.5 GB | |
| 3 | Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 35.8 t/s | 35.8 tok/s | Q4_K_M · 6.6 GB | |
| 4 | Gemma 3 12B ITGoogle Q4_K_M · 10.9 GB · 28.4 t/s | 28.4 tok/s | Q4_K_M · 10.9 GB | |
| 5 | Q4_K_M · 6.5 GB · 36.7 t/s | 36.7 tok/s | Q4_K_M · 6.5 GB |
See all 327 models on the Arc Pro B60 →Best quant that fits · speeds estimated on this card
The Arc Pro B60 runs 14B and 35B MoE models well. Xe2 (Battlemage), BMG-G21 generation, released 2025.
GenerationXe2 (Battlemage), BMG-G21released 2025
Arc Pro B60 strengths and weaknesses
Strengths
- Runs 35B MoE models: Qwen3.6 35B A3B at 48.3 tok/s in 22 GB
- Reads long prompts fast: A 4K-token prompt on Qwen3.6 35B A3B starts answering in 2.2 s
Weaknesses
- 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 24 GB here
- Slow on large dense models: Qwen3.8 27B at 13.1 tok/s
- oneAPI, not CUDA: No FP8 or NVFP4 weights and no vLLM AWQ or GPTQ; use GGUF builds
Output speed depends on memory bandwidth (456 GB/s). Grey rows need more than 24 GB.
- 9BQwen3.5 9B · Q4_K_M · 6.6 GB35.8tok/s
- 14BQwen3 14B · Q4_K_M · 11.9 GB24.0tok/s
- 27–31BQwen3.8 27B · Q4_K_M · 17.9 GB13.1tok/s
- 35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB48.3tok/s
- 70BLlama 3.3 70B Instruct · not at 4-bitneeds 51.3 GB
- 120B MoEGPT OSS 120B · not at 4-bitneeds 70.8 GB
Estimated speeds: Q4_K_M, 4K context, one user. How we estimate
All 1141 models on the Arc Pro B60
Featured onlyBest quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works
| Model | Quant | VRAM | Speed | Context | Score | Fit |
|---|---|---|---|---|---|---|
Supra 50M ReasoningChatReasoning | Q4_K_M | 0.3 GB | 633.3 t/s | 1K | S100, Runs great | |
Qwen3 14BChatFunctions | Q4_K_M | 11.8 GB | 24.0 t/s | 41K | B64, Decent | |
Nanbeige4.1 3BChat | Q4_K_M | 3.7 GB | 81.7 t/s | 262K | S88, Runs great | |
Opt 125MChat | FP16 | 0.3 GB | 814.3 t/s | 2K | S100, Runs great | |
| BF16 | 0.9 GB | 380.0 t/s | 33K | S100, Runs great | ||
Surya Ocr 2Vision | Q4_K_M | 0.8 GB | 300.0 t/s | 262K | S100, Runs great | |
| BF16 | 2.7 GB | 93.4 t/s | 41K | S90, Runs great | ||
Tiny Random Gpt2Chat | Q4_K_M | 0.0 GB | 22800.0 t/s | 1K | S100, Runs great | |
Kimi Linear Tiny RandomChatFunctions | BF16 | 0.3 GB | 459.5 t/s | — | S100, Runs great | |
| Q4_K_M | 0.3 GB | 760.0 t/s | 4K | S100, Runs great | ||
| Q4_K_M | 15.0 GB | 48.9 t/s | 203K | A79, Runs well | ||
| Q4_K_M | 21.5 GB | 108.5 t/s | 262K | A79, Runs well | ||
Nanowhale 100MChat | BF16 | 0.5 GB | 273.8 t/s | 2K | S100, Runs great | |
Seed OSS 36B InstructChatFunctions | Q4_K_S | 22.8 GB | 10.9 t/s | 524K | D27, Barely runs | |
Kimi VL A3B InstructVisionFunctions | Q4_K_M | 10.7 GB | 67.0 t/s | 131K | S85, Runs great |
Specs
- Architecture
- Xe2 (Battlemage), BMG-G21
- VRAM
- 24 GB GDDR6
- Memory bandwidth
- 456.0 GB/s
- FP16 (dense)
- 98.50 TFLOPS
- TDP
- 200 W
- Released
- 2025-09-19
- MSRP
- $599
Arc Pro B60 FAQ
What is the largest model the Arc Pro B60 can run?
Qwen3.6 35B A3B (36.0B) is the largest at 4-bit or better: at Q4_K_M it uses 22.0 GB of 24 GB, and runs at about 48.3 tok/s.
Can the Arc Pro B60 run 70B models?
No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 24 GB on this card.
How many tokens per second does the Arc Pro B60 give?
About 48.3 tok/s on Qwen3.6 35B A3B at Q4_K_M, with the first token after 2.2 s on a 4K-token prompt.
What power supply does the Arc Pro B60 need?
Plan a 400 W PSU (2× the 200 W TDP).
Is 24 GB enough for coding and agents?
Yes for coding: Qwen3.6 35B A3B, the best of the 4 coding models that fit, runs at about 48 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.