From a2c59636f596a11211bb1a49781f463b3a278136 Mon Sep 17 00:00:00 2001 From: Pedro Cuenca Date: Mon, 27 Jul 2026 10:31:52 +0200 Subject: [PATCH] Graph --- src/models/onyx.cpp | 138 +++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 136 insertions(+), 2 deletions(-) diff --git a/src/models/onyx.cpp b/src/models/onyx.cpp index 55960d8c58..dc99851179 100644 --- a/src/models/onyx.cpp +++ b/src/models/onyx.cpp @@ -55,9 +55,143 @@ void llama_model_onyx::load_arch_tensors(llama_model_loader &) { } } -llama_model_onyx::graph::graph(const llama_model & /*model*/, const llm_graph_params & params) +llama_model_onyx::graph::graph(const llama_model & model, const llm_graph_params & params) : llm_graph_context(params) { - GGML_ABORT("onyx: build_arch_graph not implemented yet"); + const int64_t n_embd_head = hparams.n_embd_head_v(); + GGML_ASSERT(n_embd_head == hparams.n_embd_head_k()); + + ggml_tensor * cur; + ggml_tensor * inpL; + + inpL = build_inp_embd(model.tok_embd); + + // OnyxNormalizedEmbedding applies scaleless RMSNorm to token embeddings + inpL = build_norm(inpL, NULL, NULL, LLM_NORM_RMS, -1); + cb(inpL, "inp_embd_normed", -1); + + ggml_tensor * inp_pos = build_inp_pos(); + auto * inp_attn = build_attn_inp_kv_iswa(); + ggml_tensor * inp_out_ids = build_inp_out_ids(); + + const float kq_scale = 1.0f / sqrtf(float(n_embd_head)); + + for (int il = 0; il < n_layer; ++il) { + const float freq_base_l = model.get_rope_freq_base (cparams, il); + const float freq_scale_l = model.get_rope_freq_scale(cparams, il); + + ggml_tensor * inpSA = inpL; + + const bool use_rope = hparams.n_no_rope_layer_step > 0 && + (il + 1) % hparams.n_no_rope_layer_step != 0; + + // pre-attention norm (weight+1 folded at conversion time) + cur = build_norm(inpL, model.layers[il].attn_norm, NULL, LLM_NORM_RMS, il); + cb(cur, "attn_norm", il); + + // self-attention: attention output gate around SDPA (afmoe.cpp:147-191) + { + ggml_tensor * attn_inp = cur; // save input for gate computation + + auto [Qcur, Kcur, Vcur] = build_qkv(model.layers[il], cur, + n_embd_head, n_head, n_head_kv, il); + + // gate = wqkv_gate @ attn_inp (from pre-attn hidden state) + ggml_tensor * gate = build_lora_mm(model.layers[il].wqkv_gate, attn_inp); + cb(gate, "attn_gate_proj", il); + + // QK-norm. attn_q_norm weight was synthesized at conversion to broadcast + // qk_scale_factor across head_dim; attn_k_norm is identity (ones). + Qcur = build_norm(Qcur, model.layers[il].attn_q_norm, NULL, LLM_NORM_RMS, il); + Kcur = build_norm(Kcur, model.layers[il].attn_k_norm, NULL, LLM_NORM_RMS, il); + cb(Qcur, "Qcur_normed", il); + cb(Kcur, "Kcur_normed", il); + + if (use_rope) { + Qcur = ggml_rope_ext( + ctx0, Qcur, inp_pos, nullptr, + n_rot, rope_type, n_ctx_orig, freq_base_l, freq_scale_l, + ext_factor, attn_factor, beta_fast, beta_slow); + cb(Qcur, "Qcur_rope", il); + + Kcur = ggml_rope_ext( + ctx0, Kcur, inp_pos, nullptr, + n_rot, rope_type, n_ctx_orig, freq_base_l, freq_scale_l, + ext_factor, attn_factor, beta_fast, beta_slow); + cb(Kcur, "Kcur_rope", il); + } + + // SDPA. wo is deferred; the gate goes between attn_out and o_proj. + cur = build_attn(inp_attn, + NULL, NULL, NULL, + Qcur, Kcur, Vcur, nullptr, nullptr, nullptr, kq_scale, il); + cb(cur, "attn_out", il); + + gate = ggml_sigmoid(ctx0, gate); + cb(gate, "attn_gate_sig", il); + cur = ggml_mul(ctx0, cur, gate); + cb(cur, "attn_gated", il); + + cur = build_lora_mm(model.layers[il].wo, cur, model.layers[il].wo_s); + cb(cur, "attn_o_proj", il); + } + + // post-attention norm + cur = build_norm(cur, model.layers[il].attn_post_norm, NULL, LLM_NORM_RMS, il); + cb(cur, "attn_post_norm", il); + + if (il == n_layer - 1 && inp_out_ids) { + cur = ggml_get_rows(ctx0, cur, inp_out_ids); + inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids); + } + + ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA); + cb(ffn_inp, "ffn_inp", il); + + // pre-FFN norm + cur = build_norm(ffn_inp, model.layers[il].ffn_norm, NULL, LLM_NORM_RMS, il); + cb(cur, "ffn_norm", il); + + // SwiGLU dense FFN + cur = build_ffn(cur, + model.layers[il].ffn_up, NULL, NULL, + model.layers[il].ffn_gate, NULL, NULL, + model.layers[il].ffn_down, NULL, NULL, + NULL, + LLM_FFN_SILU, LLM_FFN_PAR, il); + cb(cur, "ffn_out", il); + + // post-FFN norm + cur = build_norm(cur, model.layers[il].ffn_post_norm, NULL, LLM_NORM_RMS, il); + cb(cur, "ffn_post_norm", il); + + cur = ggml_add(ctx0, cur, ffn_inp); + cur = build_cvec(cur, il); + cb(cur, "l_out", il); + + inpL = cur; + } + + cur = inpL; + + // final norm + cur = build_norm(cur, model.output_norm, NULL, LLM_NORM_RMS, -1); + cb(cur, "result_norm", -1); + res->t_embd = cur; + + // lm_head. output_multiplier is already folded into the weight at conversion. + cur = build_lora_mm(model.output, cur, model.output_s); + + // Final logit tanh softcap (from gemma3.cpp). + if (hparams.f_final_logit_softcapping) { + cur = ggml_scale(ctx0, cur, 1.0f / hparams.f_final_logit_softcapping); + cur = ggml_tanh(ctx0, cur); + cur = ggml_scale(ctx0, cur, hparams.f_final_logit_softcapping); + } + + cb(cur, "result_output", -1); + res->t_logits = cur; + + ggml_build_forward_expand(gf, cur); } std::unique_ptr llama_model_onyx::build_arch_graph(const llm_graph_params & params) const {