EV

EVA Meissa big pro v2 i1

EvaGGUF by mradermacherUpdated 2024-11-25
Download on Hugging Face
Parameters21.9B
Context64K
Layers74
Architectureqwen2
Smallest quant5.1 GB
Quick answer

EVA Meissa big pro v2 i1 needs about 16.3 GB of memory at Q4_K_M (a 13.3 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 21 quants, from 5.1 GB (IQ1_S) to 18.0 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 9.9 GB.

EVA Meissa big pro v2 i1 is a 21.9B-parameter Eva model in GGUF format, quantized by mradermacher. Its trained context length is 64K tokens. The repository has 21 quant files, from 5.1 GB (IQ1_S) to 18.0 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S5.1 GB 8.1 GB
Too big Get file →
IQ1_M5.5 GB 8.5 GB
Too big Get file →
IQ2_XXS6.2 GB 9.2 GB
Too big Get file →
IQ2_XS6.8 GB 9.8 GB
Too big Get file →
IQ2_S7.2 GB 10.2 GB
Too big Get file →
IQ2_M7.8 GB 10.8 GB
Too big Get file →
Q2_K8.4 GB 11.4 GB
Too big Get file →
IQ3_XXS8.7 GB 11.7 GB
Too big Get file →
IQ3_XS9.3 GB 12.3 GB
Too big Get file →
Q3_K_S9.7 GB 12.7 GB
Too big Get file →
IQ3_S9.8 GB 12.8 GB
Too big Get file →
IQ3_M10.1 GB 13.1 GB
Too big Get file →
Q3_K_M10.8 GB 13.8 GB
Too big Get file →
Q3_K_L11.7 GB 14.7 GB
Too big Get file →
IQ4_XS11.9 GB 14.9 GB
Too big Get file →
Q4_012.6 GB 15.6 GB
Too big Get file →
Q4_K_S12.6 GB 15.6 GB
Too big Get file →
Q4_K_M13.3 GB 16.3 GB
Too big Get file →
Q5_K_S15.2 GB 18.2 GB
Too big Get file →
Q5_K_M15.6 GB 18.5 GB
Too big Get file →
Q6_K18.0 GB 21.0 GB
Too big Get file →

What GPU do you need for EVA Meissa big pro v2 i1?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: the best quant that fits is IQ3_XXS (8.7 GB file, about 11.7 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_S (12.6 GB file, about 15.6 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (18.0 GB file, about 21.0 GB in use).

Which EVA Meissa big pro v2 i1 quant should you choose?

Q4_K_M is the usual starting point for EVA Meissa big pro v2 i1: it balances file size and quality well. Here it is a 13.3 GB file.

If memory is tight, the smallest file is IQ1_S at 5.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen2
Parameters21.9B
Layers74
Trained context65,536 tokens
KV cache (fp16)296 KB per token

Source

Quantized bymradermacher
FamilyEva
Downloads (HF)142
Last updated2024-11-25

Frequently asked questions about EVA Meissa big pro v2 i1

How much VRAM does EVA Meissa big pro v2 i1 need?

About 16.3 GB at Q4_K_M with an 8K context window. That is the 13.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run EVA Meissa big pro v2 i1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 8.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of EVA Meissa big pro v2 i1 GGUF?

It depends on the quant: IQ1_S is 5.1 GB, IQ1_M is 5.5 GB, IQ2_XXS is 6.2 GB, IQ2_XS is 6.8 GB, IQ2_S is 7.2 GB, IQ2_M is 7.8 GB and 15 more.

What context length does EVA Meissa big pro v2 i1 support?

The GGUF header lists a trained context length of 65,536 tokens (64K). The KV cache grows with the context you use: about 2.5 GB at 8K and 9.9 GB at 32K.

Where can I download EVA Meissa big pro v2 i1 GGUF?

The files are on Hugging Face in the mradermacher/EVA-Meissa-big-pro-v2-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Other GGUFs of EVA-Meissa-big-pro-v2

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Eva models

All Eva →

Other medium models

All medium models →