CI Test Models

nm-testing 's Collections

CI Test Models

KV Cache Quantization

Models in CI

FP8-Block Quantized Models

LLM Compressor testing

Speculators testing

Sparse-Llama-3.1-8B-2of4

SparseGPT LLMs

FP8 Models

updated about 5 hours ago

Models used in CI tests across llm-compressor, compressed-tensors, and vllm repositories.

Upvote

nm-testing/Llama-3.2-1B-Instruct-FP8-KV

1B • Updated Nov 1, 2024 • 36.4k
nm-testing/Llama-3.2-1B-Instruct-quip-w4a16

2B • Updated Sep 12, 2025 • 8.34k
nm-testing/Llama-3.2-1B-Instruct-spinquantR1R2R4-w4a16

2B • Updated Aug 22, 2025 • 8.48k
nm-testing/Llama3_2_1B_speculator.eagle3

0.4B • Updated Nov 15, 2025 • 128k
nm-testing/Meta-Llama-3-8B-FP8-compressed-tensors-test

Text Generation • 8B • Updated Oct 9, 2024 • 21.9k
nm-testing/Meta-Llama-3-8B-Instruct-W8A8-Static-Per-Tensor-Asym

8B • Updated Dec 11, 2024 • 106
nm-testing/Meta-Llama-3-8B-Instruct-W8A8-Static-Per-Tensor-Sym

8B • Updated Dec 10, 2024 • 112
nm-testing/Meta-Llama-3-8B-Instruct-nonuniform-test

Text Generation • 8B • Updated Oct 9, 2024 • 41.8k
nm-testing/Mixtral-8x7B-Instruct-v0.1-W4A16-channel-quantized

47B • Updated Oct 9, 2024 • 3
nm-testing/Mixtral-8x7B-Instruct-v0.1-W4A16-quantized

47B • Updated Oct 9, 2024 • 92
nm-testing/Mixtral-8x7B-Instruct-v0.1-W8A16-quantized

47B • Updated Oct 9, 2024 • 6
nm-testing/Phi-3-mini-128k-instruct-FP8

4B • Updated Oct 9, 2024 • 1.45k
nm-testing/Qwen1.5-MoE-A2.7B-Chat-quantized.w4a16

14B • Updated Feb 24, 2025 • 121k • 1
nm-testing/Qwen2-1.5B-Instruct-FP8W8

Text Generation • 2B • Updated Oct 9, 2024 • 6.51k
nm-testing/Qwen3-30B-A3B-FP8-block

Text Generation • 3B • Updated Oct 27, 2025 • 6.39k
nm-testing/Qwen3-30B-A3B-Fp8-v1

31B • Updated Jun 4, 2025 • 6.47k
nm-testing/Qwen3-30B-A3B-MXFP4A16

17B • Updated Feb 17 • 21k
nm-testing/Qwen3-4B-mixed-quant-RTN-wNaM

4B • Updated 9 days ago • 2.18k
nm-testing/Qwen3-Next-80B-A3B-Instruct-NVFP4

Updated Sep 16, 2025 • 2.4k • 2
nm-testing/SmolLM-1.7B-Instruct-quantized.w4a16

Text Generation • 2B • Updated 7 days ago • 1.44M
nm-testing/Speculator-Qwen3-30B-MOE-VL-Eagle3

0.4B • Updated Jan 22 • 411
nm-testing/Speculator-Qwen3-8B-Eagle3-converted-071-quantized

1B • Updated Jul 29, 2025 • 21.9k
nm-testing/Speculator-Qwen3-8B-Eagle3-converted-071-quantized-w4a16

1B • Updated Sep 29, 2025 • 9.77k
nm-testing/SpeculatorLlama3-1-8B-Eagle3-converted-0717-quantized

1.0B • Updated Jul 29, 2025 • 22.6k
nm-testing/TinyLlama-1.1B-Chat-v1.0-FP8-Dynamic-compressed

1B • Updated Jan 14, 2025 • 1.12k
nm-testing/TinyLlama-1.1B-Chat-v1.0-FP8-Dynamic-uncompressed

1B • Updated Jan 14, 2025 • 234
nm-testing/TinyLlama-1.1B-Chat-v1.0-MXFP4

0.6B • Updated May 5 • 8.28k
nm-testing/TinyLlama-1.1B-Chat-v1.0-NVFP4

0.7B • Updated May 12 • 25.8k
nm-testing/TinyLlama-1.1B-Chat-v1.0-NVFP4A16

0.7B • Updated May 12 • 10.7k
nm-testing/TinyLlama-1.1B-Chat-v1.0-W4A16-G128-Asym-Updated-ActOrder

1B • Updated Apr 10, 2025 • 16.3k
nm-testing/TinyLlama-1.1B-Chat-v1.0-W4A16-G128-compressed

1B • Updated Jan 14, 2025 • 857
nm-testing/TinyLlama-1.1B-Chat-v1.0-W4A16-G128-uncompressed

1B • Updated Jan 14, 2025 • 222
nm-testing/TinyLlama-1.1B-Chat-v1.0-W8A16-G128-compressed

1B • Updated Jan 14, 2025 • 709
nm-testing/TinyLlama-1.1B-Chat-v1.0-W8A16-G128-uncompressed

1B • Updated Jan 14, 2025 • 238
nm-testing/TinyLlama-1.1B-Chat-v1.0-W8A8-Dynamic-Per-Token-compressed

1B • Updated Jan 14, 2025 • 688
nm-testing/TinyLlama-1.1B-Chat-v1.0-W8A8-Dynamic-Per-Token-uncompressed

1B • Updated Jan 14, 2025 • 232
nm-testing/TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-attn_head

1B • Updated Jan 14 • 2.37k
nm-testing/TinyLlama-1.1B-Chat-v1.0-kvcache-fp8-tensor

1B • Updated Jan 14 • 8.28k
nm-testing/asym-w8w8-int8-static-per-tensor-tiny-llama

1B • Updated Oct 9, 2024 • 8.65k
nm-testing/dflash-qwen3-8b-speculators

2B • Updated Apr 15 • 24.9k
nm-testing/llama2.c-stories15M

Text Generation • 24.4M • Updated Mar 10, 2025 • 6.76k
nm-testing/llama2.c-stories15M-ultrachat-mixed-compressed

15.2M • Updated May 12, 2025 • 1.31k
nm-testing/llama2.c-stories15M-ultrachat-mixed-uncompressed

24.4M • Updated May 12, 2025 • 1.73k
nm-testing/llama2.c-stories42M-gsm8k-quantized-only-compressed

58.1M • Updated Feb 12, 2025 • 2.3k
nm-testing/llama2.c-stories42M-gsm8k-quantized-only-uncompressed

58.2M • Updated Feb 12, 2025 • 2.03k
nm-testing/pixtral-12b-FP8-dynamic

Image-Text-to-Text • 13B • Updated Apr 11, 2025 • 1.72k • 1
nm-testing/qwen3-8b-peagle-speculators

2B • Updated May 7 • 34.6k
nm-testing/test-w4a16-mixtral-actorder-group

47B • Updated Dec 26, 2024 • 1.46k
nm-testing/tinyllama-fp8-dynamic-compressed

1B • Updated Oct 9, 2024 • 766
nm-testing/tinyllama-oneshot-w4a16-channel-v2

Text Generation • 1B • Updated Oct 9, 2024 • 22.9k • 1
nm-testing/tinyllama-oneshot-w4a16-group128-v2

Text Generation • 1B • Updated Oct 9, 2024 • 17.9k
nm-testing/tinyllama-oneshot-w8-channel-a8-tensor

Text Generation • 1B • Updated Oct 9, 2024 • 1.17k
nm-testing/tinyllama-oneshot-w8a16-per-channel

Text Generation • 1B • Updated Oct 9, 2024 • 1.05k
nm-testing/tinyllama-oneshot-w8a8-channel-dynamic-token-v2

Text Generation • 1B • Updated Oct 9, 2024 • 15.8k
nm-testing/tinyllama-oneshot-w8a8-dynamic-token-v2

Text Generation • 1B • Updated Oct 9, 2024 • 13.3k
nm-testing/tinyllama-oneshot-w8w8-test-static-shape-change

Text Generation • 1B • Updated Oct 9, 2024 • 65.6k
nm-testing/tinyllama-w4a16-compressed

1B • Updated Oct 9, 2024 • 1.64k
nm-testing/tinyllama-w8a16-dense

1B • Updated Mar 7 • 735
nm-testing/tinyllama-w8a8-compressed

1B • Updated Oct 9, 2024 • 1.2k
nm-testing/tinysmokellama-3.2

354k • Updated Jun 12 • 161k
nm-testing/tinysmokeqwen3

2.64M • Updated Jun 15 • 38.8k
nm-testing/tinysmokeqwen3moe

2.93M • Updated Nov 20, 2025 • 13.2k
nm-testing/tinysmokeqwen3moe-W4A16-first-only

2.93M • Updated Nov 20, 2025 • 2
nm-testing/tinysmokeqwen3moe-W4A16-first-only-CTstable

2.93M • Updated Nov 25, 2025 • 17.5k

Upvote

Collection guide
Browse collections