vulkan: add TQ2_0 (ternary) support (#25850)
* vulkan: TQ2_0 (ternary) support — dequant + dedicated mul_mat_vec + matmul via dequant_funcs First Vulkan ternary type in ggml. Correctness: OM-125m TQ2_0 vs F16 top-12 logprobs identical to 4 decimals fully offloaded (float dequant path, no Q8_K activation quant). Speed at 125m ~= F16 (overhead-bound at this scale); the bandwidth win targets larger BitNet SKUs. MMQ/int-dot path intentionally not wired yet. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * tests: enable TQ2_0 in backend-ops type lists Vulkan now implements TQ2_0 (dequant, mul_mat_vec, mul_mm, get_rows); backends without support skip via not-supported as usual. TQ1_0 stays disabled. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Michael Trabalka <michael.trabalka@sqv.ai> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Fable 5
Michael Trabalka
parent
55f453b924
commit
a7cd2f0e98
@@ -608,6 +608,20 @@ vec2 get_dm(uint ib, uint a_offset) {
|
||||
}
|
||||
#endif
|
||||
|
||||
#if defined(DATA_A_TQ2_0)
|
||||
vec2 dequantize(uint ib, uint iqs, uint a_offset) {
|
||||
// elem e -> byte qs[(e/128)*32 + e%32], bits 2*((e%128)/32); w = q - 1 (d applied via get_dm)
|
||||
const uint qsi = (iqs / 128) * 32 + (iqs % 32); // iqs even -> qsi, qsi+1 in same group/level
|
||||
const uint shift = 2 * ((iqs % 128) / 32);
|
||||
|
||||
const uvec2 qs = uvec2(data_a[a_offset + ib].qs[qsi], data_a[a_offset + ib].qs[qsi + 1]);
|
||||
return vec2((qs >> shift) & 3) - 1.0;
|
||||
}
|
||||
vec2 get_dm(uint ib, uint a_offset) {
|
||||
return vec2(float(data_a[a_offset + ib].d), 0);
|
||||
}
|
||||
#endif
|
||||
|
||||
#if defined(DATA_A_Q3_K)
|
||||
vec2 dequantize(uint ib, uint iqs, uint a_offset) {
|
||||
iqs /= 2;
|
||||
|
||||
Reference in New Issue
Block a user