d1-3B decision model in GGUF, compressed by TurinTech with Artemis. Use the /v1/systemone endpoint, not chat. Recommended: d1-3B-Artemis-Q4_K_M.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/d1-3B-Artemis-GGUF. Download size is not runtime memory. Measured Q4_K_M: same answer as BF16 93.75%, 0 confident answers changed on the 80-case final set; protocol rtx3060-1e6f04a-20261023.
d1-3B decision model in GGUF, compressed by TurinTech with Artemis. Use the /v1/systemone endpoint, not chat. Recommended: d1-3B-Artemis-Q4_K_M.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/d1-3B-Artemis-GGUF. Download size is not runtime memory. Measured Q4_K_M: same answer as BF16 93.75%, 0 confident answers changed on the 80-case final set; protocol rtx3060-1e6f04a-20261023.
d1-3B decision model in GGUF, compressed by TurinTech with Artemis. Use the /v1/systemone endpoint, not chat. Recommended: d1-3B-Artemis-Q4_K_M.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/d1-3B-Artemis-GGUF. Download size is not runtime memory. Measured Q4_K_M: same answer as BF16 93.75%, 0 confident answers changed on the 80-case final set; protocol rtx3060-1e6f04a-20261023.
d1-3B decision model in GGUF, compressed by TurinTech with Artemis. Use the /v1/systemone endpoint, not chat. Recommended: d1-3B-Artemis-Q4_K_M.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/d1-3B-Artemis-GGUF. Download size is not runtime memory. Measured Q4_K_M: same answer as BF16 93.75%, 0 confident answers changed on the 80-case final set; protocol rtx3060-1e6f04a-20261023.