Default Branch

b2cb4512c5 · Create parameters overview (#1269) · Updated 2026-02-20 07:20:56 +01:00

Branches

4b6f0ff9c1 · q2_K · Updated 2025-06-18 12:58:31 +02:00    git

4211
3761

b7744eee27 · iq2_ks · Updated 2025-06-17 16:26:13 +02:00    git

4211
3759

72fd9faa9f · Slightly faster · Updated 2025-06-16 09:43:44 +02:00    git

4211
3772

8d97f53699 · Call iqk_convert_repack in MoE GEMM · Updated 2025-06-14 04:47:00 +02:00    git

4211
3749

7986500f9d · Add all IQK quants · Updated 2025-06-12 19:25:43 +02:00    git

4211
3747

8ba852a2f3 · Remove the scales, they are not needed · Updated 2025-06-12 18:26:42 +02:00    git

4211
3749

1a8a0e5e63 · Perhaps a slightly better version for IQ2_XXS, IQ3_XXS, IQ3_S GEMV · Updated 2025-06-12 18:09:31 +02:00    git

4211
3745

cdcb324fe6 · Better strategy for GPU offload · Updated 2025-06-11 18:44:05 +02:00    git

4211
3743

ec530d4e5f · iq3_s: much faster GEMM via repacking to q8_0_r8 · Updated 2025-06-11 15:19:54 +02:00    git

4211
3743

3d5672073f · Faster iq1_s GEMM via repacking to Q8_0_R8 · Updated 2025-06-11 13:38:48 +02:00    git

4211
3742

be3b768c9a · Much faster iq3_xxs GEMM via repacking to q8_0_r8 (AVX2) · Updated 2025-06-11 11:45:59 +02:00    git

4211
3741

415a7cf6c3 · NEON is not working yet, so still use Q8_K GEMM · Updated 2025-06-11 09:55:42 +02:00    git

4211
3744

c8cf128099 · Add missing break · Updated 2025-06-09 14:18:11 +02:00    git

4211
3754

a1ff316378 · update XTC docs · Updated 2025-06-09 12:12:24 +02:00    git

4211
3739

f59fe11764 · Adding forgottent file · Updated 2025-06-09 08:59:39 +02:00    git

4211
3745

df600c1ec0 · Add list_saved_prompts function to server · Updated 2025-06-06 19:23:07 +02:00    git

4211
3731

7139bebd12 · Fix #499 · Updated 2025-06-06 18:30:15 +02:00    git

4211
3729

a8f1b74a5f · Playing · Updated 2025-06-06 14:22:02 +02:00    git

4211
3730

fdfad30721 · Just leave the check. · Updated 2025-06-06 11:05:10 +02:00    git

4211
3730

b1ad163155 · Merge remote-tracking branch 'origin/main' into s6/MLA_prompt_save_restore_fix · Updated 2025-06-06 10:22:29 +02:00    git

4211
3735