Hemmingway-1 27B VRAM requirements

Hemmingway-1 27B needs about 18.0 GB of VRAM at Q4_K_M with 8K context, so it fits a 24 GB card such as the RTX 3090 or the RTX 4090. Hemmingway-1 27B has 27.3B parameters. With the same 8K context and one request it needs 29.0 GB at FP8 and 57.0 GB at FP16/BF16. The published weights take 50.9 GB (BF16). The smallest setup here that holds it at Q4_K_M with an 8K context is the RTX 3090 (24 GB), which runs it with up to 88K tokens of context.

18.0 GB at Q4_K_M, 8K context, one request

FP8
29.0 GB
FP16 / BF16
57.0 GB
Published weights
50.9 GB
Smallest setup, Q4_K_M
RTX 3090 (24 GB)
Open Hemmingway-1 27B in the calculator

Worked example: Hemmingway-1 27B with 32K tokens

Inputs: Hemmingway-1 27B · Q4_K_M weights · 32,768 tokens of context · one request · FP16 KV cache

Weights (27.3B at Q4_K_M)
15.4 GB
KV cache (64 KB per token)
2.00 GB
Buffers and runtime (0.5 GB + 10%)
2.24 GB
Total
19.6 GB

The smallest setup here that holds it is the RTX 3090 (24 GB), with about 4.37 GB to spare. Change the inputs in the calculator

What makes Hemmingway-1 27B's memory use different

48 of its 64 layers keep a fixed-size linear-attention state, so the cache grows by 64 KB per token where caching every layer would take 256 KB.

Per token of context it adds 64 KB of FP16 cache; Granite 4.2 30B (29.3B), the nearest-sized model here with plain full attention, adds 256 KB, so Hemmingway-1 27B needs 25% as much.

How much VRAM does Hemmingway-1 27B need?

Hemmingway-1 27B needs 18.0 GB at Q4_K_M, 30.8 GB at Q8_0 and 57.0 GB at FP16 with 8,192 tokens of context and one request. Each total below is the weights plus the FP16 KV cache and the runtime overhead (0.5 GB plus 10%), and each row opens the calculator with that setting. What the GGUF names mean.

PrecisionWeightsTotalSmallest setup
As published (BF16) 50.9 GB 57.0 GB 2× RTX 5090 (64 GB)
FP16 / BF16 50.9 GB 57.0 GB 2× RTX 5090 (64 GB)
FP8 / INT8 25.4 GB 29.0 GB RTX 5090 (32 GB)
INT4 (AWQ / GPTQ) 13.5 GB 15.9 GB RTX 4060 Ti 16GB
GGUF Q8_0 27.0 GB 30.8 GB RTX 5090 (32 GB)
GGUF Q6_K 20.9 GB 24.0 GB RTX 5090 (32 GB)
GGUF Q5_K_M 18.0 GB 20.9 GB RTX 3090 (24 GB)
GGUF Q4_K_M 15.4 GB 18.0 GB RTX 3090 (24 GB)
GGUF IQ4_XS 13.8 GB 16.3 GB RTX 3090 (24 GB)
GGUF Q3_K_M 12.4 GB 14.7 GB RTX 4060 Ti 16GB
GGUF IQ3_XXS 10.5 GB 12.6 GB RTX 4060 Ti 16GB
GGUF Q2_K 10.7 GB 12.8 GB RTX 4060 Ti 16GB

Fine-tuning Hemmingway-1 27B? Hemmingway-1 27B VRAM for LoRA, QLoRA and full training.

Hemmingway-1 27B GGUF files on Hugging Face

The 8 GGUF files llama.cpp picks for Hemmingway-1 27B with -hf <repo>:<QUANT>, with the bytes the Hugging Face file list reported on 2026-09-29. Each total is that file plus the FP16 KV cache and 0.5 GB + 10%, one request. "vs estimate" compares the file with the bits-per-weight size the table above uses.

QuantFileSizevs estimate 8K32K128K
Q2_K bartowski/Altworld_Hemmingway-1-GGUF 10.1 GB10,821,544,000 bytes −5% 12.1 GB13.8 GB20.4 GB
Q3_K_M bartowski/Altworld_Hemmingway-1-GGUF 12.5 GB13,404,051,520 bytes ±0% 14.8 GB16.4 GB23.0 GB
IQ4_XS bartowski/Altworld_Hemmingway-1-GGUF 14.4 GB15,475,951,680 bytes +4% 16.9 GB18.6 GB25.2 GB
Q4_K_M bartowski/Altworld_Hemmingway-1-GGUF 16.2 GB17,442,400,320 bytes +6% 18.9 GB20.6 GB27.2 GB
Q5_K_M bartowski/Altworld_Hemmingway-1-GGUF 19.5 GB20,923,877,440 bytes +8% 22.5 GB24.1 GB30.7 GB
Q6_K bartowski/Altworld_Hemmingway-1-GGUF 22.2 GB23,860,566,080 bytes +7% 25.5 GB27.1 GB33.7 GB
Q8_0 bartowski/Altworld_Hemmingway-1-GGUF 27.1 GB29,116,389,440 bytes ±0% 30.9 GB32.5 GB39.1 GB
BF16 bartowski/Altworld_Hemmingway-1-GGUF2 parts 50.9 GB54,657,734,624 bytes ±0% 57.0 GB58.7 GB65.3 GB

GB here is GiB (1024³ bytes), as everywhere on this site; Hugging Face shows the Q4_K_M file as 17.4 GB, in 10⁹ bytes.

Which Hemmingway-1 27B GGUF fits a 16, 24, 32, 48 or 80 GB GPU or a Mac

From the file sizes above: the largest file that fits each machine with an 8K context and 0.5 GB left free, the longest context it then has room for, and the longest context for Q4_K_M (16.2 GB). One request, FP16 KV cache.

HardwareLargest GGUF, 8KIts longest contextQ4_K_M, longest context
RTX 5060 Ti 16GB Q3_K_M, 12.5 GB 18K No
RTX 4090 (24 GB) Q5_K_M, 19.5 GB 22K 74K
RTX 5090 (32 GB) Q8_0, 27.1 GB 16K 190K
L40S (48 GB) Q8_0, 27.1 GB 249K 256K (full)
H100 SXM (80 GB) BF16, 50.9 GB 256K (full) 256K (full)
M4 Pro Mac (64 GB, 48 GB usable) Q8_0, 27.1 GB 249K 256K (full)
M4 Max Mac (128 GB, 96 GB usable) BF16, 50.9 GB 256K (full) 256K (full)

Run Hemmingway-1 27B with llama.cpp

llama-server -hf bartowski/Altworld_Hemmingway-1-GGUF:Q4_K_M -c 75776

Altworld_Hemmingway-1-Q4_K_M.gguf, 17.4 GB, from bartowski/Altworld_Hemmingway-1-GGUF (checked 2026-09-29). On a 24 GB RTX 3090 / 4090, Q4_K_M with -c 75776 takes 23.5 GB and leaves 557 MB. The file is 871 MB over the Q4_K_M estimate, so -c counts the file.

Hemmingway-1 27B at 8K, 32K, 128K and 256K (full) tokens of context

16 of its 64 layers use full attention and 48 are linear-attention layers with no growing cache. Each extra token of context adds 64 KB of FP16 cache per request. The last column is the smallest setup here, one card or a group, that holds it at Q4_K_M. How the KV cache works.

ContextKV cache, FP16Total at Q4_K_MTotal at Q8_0Smallest setup, Q4_K_M
8K tokens 512 MB 18.0 GB 30.8 GB RTX 3090 (24 GB)
32K tokens 2.00 GB 19.6 GB 32.4 GB RTX 3090 (24 GB)
128K tokens 8.00 GB 26.2 GB 39.0 GB RTX 5090 (32 GB)
256K tokens 16.0 GB 35.0 GB 47.8 GB M4 Pro Mac (64 GB, 48 GB usable)

How fast Hemmingway-1 27B writes

Tokens per second for one request with 8,192 tokens of context, estimated from memory bandwidth. A dash means it does not fit on one card. Try other settings in the speed calculator.

HardwareBandwidthQ4_K_MFP8
RTX 3060 12GB 360 GB/s ——
RTX 4090 1,008 GB/s 31–43—
RTX 5090 1,792 GB/s 53–7634–47
M4 Max Mac (128 GB) 546 GB/s 17–2411–15
M3 Ultra Mac Studio (512 GB) 819 GB/s 25–3516–22
H100 SXM 3,350 GB/s 93–13760–86
H200 4,800 GB/s 126–19183–121

Longest context on one GPU

How many tokens of context Hemmingway-1 27B fits on a single card with one request, an FP16 KV cache and 0.5 GB left free. "Full" means the model's whole context window fits.

GPUMemoryQ4_K_MQ8_0FP8
RTX 3060 12 GB NoNoNo
RTX 4060 Ti 16GB 16 GB NoNoNo
RTX 3090 / 4090 24 GB 88KNoNo
RTX 5090 32 GB 204K18K43K
A100 40GB 40 GB 256K (full)134K160K
Mac, 64 GB unified memory 48 GB 256K (full)251K256K (full)
L40S / RTX 6000 Ada 48 GB 256K (full)251K256K (full)
A100 / H100 80GB 80 GB 256K (full)256K (full)256K (full)
Mac, 128 GB unified memory 96 GB 256K (full)256K (full)256K (full)
H200 141 GB 256K (full)256K (full)256K (full)
B200 180 GB 256K (full)256K (full)256K (full)

Model details

Parameters
27.3B (27,320,697,856)
Layers
16 of its 64 layers use full attention and 48 are linear-attention layers with no growing cache
Attention cache
4 KV heads × 256
Context length
262,144 tokens
Published weights
50.9 GB (BF16)
On Hugging Face
Altworld/Hemmingway-1
Reported benchmarks
None published in the model card for GPQA, MMLU-Pro or LiveCodeBench.

Why this estimate looks this way

At Q4_K_M and an 8K-token context, Hemmingway-1 27B uses 15.4 GB for weights, 512 MB for its FP16 KV cache and 2.09 GB for estimated runtime overhead, totaling 18.0 GB. The overhead is a 0.5 GB base plus 10% of weights and cache.

Its attention layout matters for long context: 16 of its 64 layers use full attention and 48 are linear-attention layers with no growing cache. The FP16 cache grows by about 64 KB per additional token per request.

The architecture and published checkpoint size come from the model's config.json and weight files. These are estimates rather than measured peak memory; inference engines can reserve extra buffers or preallocate the full configured cache.

Hemmingway-1 27B next to similar-sized models

The three models here closest to it in parameter count, at Q4_K_M with 8,192 tokens of context.

ModelParametersTotalKV per tokenSmallest setup
Hemmingway-1 27B 27.3B 18.0 GB 64 KB RTX 3090 (24 GB)
ThinkingCap Qwen3.8 27B 27.8B 18.3 GB 64 KB RTX 3090 (24 GB)
Qwen3.6 27B 27.8B 18.3 GB 64 KB RTX 3090 (24 GB)
Qwen3.8 27B 27.8B 18.3 GB 64 KB RTX 3090 (24 GB)

Hemmingway-1 27B VRAM questions

How much VRAM do I need to run Hemmingway-1 27B locally?

About 18.0 GB at Q4_K_M with an 8K context and one request, by the bits-per-weight estimate, and 57.0 GB at BF16. The Q4_K_M GGUF Altworld_Hemmingway-1-Q4_K_M.gguf in bartowski/Altworld_Hemmingway-1-GGUF is 17,442,400,320 bytes (16.2 GB); with an 8K context it needs 18.9 GB. The smallest setup listed here that holds Q4_K_M is the RTX 3090 (24 GB).

Can I run Hemmingway-1 27B on a 24 GB GPU?

Yes. On the RTX 4090 (24 GB), the largest listed GGUF that fits with an 8K context and 0.5 GB free is Q5_K_M (19.5 GB), with up to 22K tokens of context. Q4_K_M (16.2 GB) runs with up to 74K tokens of context.

Which quantization of Hemmingway-1 27B fits in 16 GB of VRAM?

On the RTX 5060 Ti 16GB, the largest listed GGUF that fits with an 8K context and 0.5 GB free is Q3_K_M (12.5 GB), with up to 18K tokens of context.

Can I run Hemmingway-1 27B on a Mac?

On the M4 Pro Mac (64 GB, 48 GB usable), the largest listed GGUF that fits with an 8K context and 0.5 GB free is Q8_0 (27.1 GB), with up to 249K tokens of context. Q4_K_M (16.2 GB) runs with its full 256K-token context. On the M4 Max Mac (128 GB, 96 GB usable), the largest listed GGUF that fits with an 8K context and 0.5 GB free is BF16 (50.9 GB), with its full 256K-token context. By default macOS lets the GPU use about 75% of unified memory.

How much more VRAM does Hemmingway-1 27B need for 32K or 128K tokens of context?

Its FP16 KV cache is 512 MB at 8K, 2.00 GB at 32K, 8.00 GB at 128K for one request, so 128K adds 7.50 GB over 8K, plus 10% overhead. A q8_0 cache (-ctk q8_0 -ctv q8_0) takes 4.25 GB at 128K. Its 48 linear-attention layers keep a fixed-size state, so only 16 of 64 layers add cache.

Why does Hemmingway-1 27B use more VRAM than its GGUF file size?

The file holds only the weights. At Q4_K_M with an 8K context: 16.2 GB of weights (the file's 17,442,400,320 bytes) + 512 MB of FP16 KV cache + 2.17 GB of compute buffers and runtime (0.5 GB + 10%) = 18.9 GB.

Badge for your model card

Paste it into a Hugging Face or GitHub README; it shows the Q4_K_M total above and links to this page. Q8_0 and FP16 badges.

Hemmingway-1 27B VRAM: 18.0 GB at Q4_K_M, 8K context

Numbers read from the model files on Hugging Face on . Compare it with other models in the reproducible model and GPU report.