Where it runs 8K context · ≥ 5 t/s floor
Same plan engine as the planner, with its default reserves and RAM. How plans are chosen.
Files exact Hugging Face bytes
| Artifact | Size | Files | Revision | Source |
|---|
| Q8_0 unsloth/MiniMax-M2.5-GGUF | 226 GiB | 6 | 7c50dca0e5 | HF files |
| Q5_K_M unsloth/MiniMax-M2.5-GGUF | 151 GiB | 5 | 7c50dca0e5 | HF files |
| Q4_K_M unsloth/MiniMax-M2.5-GGUF | 129 GiB | 4 | 7c50dca0e5 | HF files |
| Native safetensors mixed/custom native; quant_method=fp8; stored tensor dtypes= MiniMaxAI/MiniMax-M2.5 | 214 GiB | 125 | f710177d93 | HF files |
Not found in the representative quant search (not proven unavailable): GGUF FP16, MLX 8-bit, MLX 4-bit, EXL3, AWQ, GPTQ. Complete selected file sets; projectors, MTP/draft and lookup files are listed separately and budgeted only by plans that use them. File bytes are not allocated memory.
Quality
No Artificial Analysis evaluation mapped to this checkpoint.
Arena text rating
1,359
CI 1355–1363 · 42,815 votes · published 2026-10-15 · dataset
Explicit name identity; exact hosted revision/local quant equivalence not verified
Publisher-reported benchmarks
None extracted from the model card.
AA and Arena describe the hosted model in the stated mode, not each quantized file; scores are never inherited by base, pruned, abliterated or quantized variants and never averaged together.
How to run it
model · EPYC 9355 system, 1511 GB total RAM; 8× RTX 5090 installed; each test TP=1
source-documented, not executed source
/home/yyj/kt_test/SGLang-f6adb4f-kt1b4665e-x86_64.AppImage --host 0.0.0.0 --port 30008 --model /mnt/data/models/MiniMax-M2.5 --kt-weight-path /mnt/data/models/MiniMax-M2.5 --kt-cpuinfer 16 --kt-threadpool-count 1 --kt-num-gpu-experts 8 --kt-method FP8 --kt-max-deferred-experts-per-token 0 --kt-expert-placement-strategy uniform --attention-backend flashinfer --fp8-gemm-backend cutlass --trust-remote-code --mem-fraction-static 0.85 --chunked-prefill-size 4096 --max-running-requests 1 --max-total-tokens 32000 --enable-mixed-chunk --tensor-parallel-size 1 --disable-shared-experts-fusion
Recipes are source examples (settings are the publisher’s or author’s, not measured optima) and were not executed by LocalFit. A downloadable file does not prove your backend build supports the architecture.
Reported speeds 5
| Hardware | Quant | Backend | Placement | Prompt t/s | Gen t/s | Date | Source |
|---|
| EPYC 9355 system, 1511 GB total RAM; 8× RTX 5090 installed; each test TP=1 | FP8 | KTransformers kt-kernel + SGLang | GPU + CPU experts | 6.8 | 15 method | 2026-04-03 | source |
| EPYC 9355 system, 1511 GB total RAM; 8× RTX 5090 installed; each test TP=1 | FP8 | KTransformers kt-kernel + SGLang | GPU + CPU experts | 146 | 16 method | 2026-04-03 | source |
| EPYC 9355 system, 1511 GB total RAM; 8× RTX 5090 installed; each test TP=1 | FP8 | KTransformers kt-kernel + SGLang | GPU + CPU experts | 204 | 16 method | 2026-04-03 | source |
| EPYC 9355 system, 1511 GB total RAM; 8× RTX 5090 installed; each test TP=1 | FP8 | KTransformers kt-kernel + SGLang | GPU + CPU experts | 317 | 16 method | 2026-04-03 | source |
| EPYC 9355 system, 1511 GB total RAM; 8× RTX 5090 installed; each test TP=1 | FP8 | KTransformers kt-kernel + SGLang | GPU + CPU experts | 318 | 16 method | 2026-04-03 | source |
Reported by the linked sources, not reproduced by LocalFit. Stream pieces/s and method-unknown rates are flagged and cannot pass a tokens/s floor. All rows