Can the Apple M3 Ultra (512 GB) run Sarvam 105B?

Yes, it runs great

Runs great. At Q6_K it needs about 119 GB of 384 GB usable, with an estimated speed of 49.4 tok/s.

Best quant: Q6_K
Generation speed: 49.4 tok/s
Prompt processing: 1,761 tok/s
First token (1K prompt): 0.6 s
Largest safe context: 128K

Best app to run it on the Apple M3 Ultra (512 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX

Point-and-click: search, download and chat in one window.

Open Sarvam 105B in LM Studio

llama.cpp Usually the first to support new model architectures.
Ollama Runs the Hugging Face GGUF with ollama run hf.co/…

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Sarvam 105B

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

huggingface.co/sarvamai/sarvam-105b-gguf (opens in a new tab)

ollama run hf.co/sarvamai/sarvam-105b-gguf

Original weights · Hugging Face

Full-precision files from Sarvam AI, for fine-tuning, vLLM, Transformers.

huggingface.co/sarvamai/sarvam-105b (opens in a new tab)

No sign-in needed.

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 74Great quality (A) · 74 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q6_K, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs great: How well it runs on the Apple M3 Ultra (512 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	33.4 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.	25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	7.1 / 15
Match score	80.5 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the Apple M3 Ultra (512 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	74.0 GB	384 GB	98.4 tok/s	Runs great	52Fair quality (C) · 52 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q3_K_M`	83.6 GB	384 GB	82.1 tok/s	Runs great	63Good quality (B) · 63 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q4_K_M`	96.3 GB	384 GB	67.0 tok/s	Runs great	71Great quality (A) · 71 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	GGUF · part 1 of 9 (opens in a new tab)
`Q5_K_M`	107 GB	384 GB	57.2 tok/s	Runs great	72Great quality (A) · 72 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q6_K`	119 GB	384 GB	49.4 tok/s	Runs great	74Great quality (A) · 74 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q8_0`	144 GB	384 GB	37.7 tok/s	Runs great	75Great quality (A) · 75 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`F16`	244 GB	384 GB	18.7 tok/s	Runs well	75Great quality (A) · 75 of 100: how capable the model is at F16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

74Great quality (A) · 74 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. Apple M2 Ultra (192 GB) Q6_K · 48.4 tok/s · Runs great
72Great quality (A) · 72 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q5_K_M · 111 tok/s · Runs great
71Great quality (A) · 71 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. Apple M4 Max (128 GB) Q4_K_M · 48.3 tok/s · Runs great
74Great quality (A) · 74 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Apple M3 Ultra (256 GB) Q6_K · 49.4 tok/s · Runs great
71Great quality (A) · 71 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX Spark N1X (6144-core GPU) (128 GB) Q4_K_M · 32.8 tok/s · Runs great
71Great quality (A) · 71 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GB10 (128 GB) Q4_K_M · 30.1 tok/s · Runs great

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.