[GGUF FE] Stateful conversion, op-translator fixes and test infrastructure

.github/components.yml

@@ -107,6 +107,12 @@ PROXY:
     - GPU
   build: []
 
+GGUF_FE:
+  revalidate:
+    - CPU # the per-op tests compile and infer their converted graphs on the CPU plugin
+  build:
+    - CPU
+
 IR_FE:
   revalidate:
     - C_API

.github/coverage/tests_cpp.yml

@@ -98,6 +98,11 @@ tests:
     mode: gtest_single
     profiles: [cpu]
 
+  - name: ov_gguf_frontend_tests
+    binary: ov_gguf_frontend_tests
+    mode: gtest_single
+    profiles: [cpu]
+
   - name: ov_inference_functional_tests
     binary: ov_inference_functional_tests
     mode: gtest_single

.github/workflows/job_cxx_unit_tests.yml

@@ -129,6 +129,7 @@ jobs:
           ${{ env.INSTALL_TEST_DIR }}/ov_ir_frontend_tests --gtest_print_time=1 --gtest_output=xml:${{ env.INSTALL_TEST_DIR }}/TEST-IRFrontend.xml
 
       - name: GGUF frontend tests
+        if: fromJSON(inputs.affected-components).GGUF_FE.test
         run: |
           ${{ env.SOURCE_COMMAND }} ${{ env.SETUPVARS }}
           ${{ env.INSTALL_TEST_DIR }}/ov_gguf_frontend_tests --gtest_print_time=1 --gtest_output=xml:${{ env.INSTALL_TEST_DIR }}/TEST-GGUFFrontend.xml

.github/workflows/ubuntu_24.yml

@@ -109,6 +109,7 @@ jobs:
       build-debian-packages: true
       build-contrib: true
       build-additional-python-packages: true
+      generate-gguf-fixtures: true
       target-branch: ${{ inputs.target-branch }}
       cmake-options: >-
             -G 'Ninja Multi-Config'

src/frontends/gguf/include/openvino/frontend/gguf/decoder.hpp

@@ -11,6 +11,7 @@
 #include <string>
 #include <vector>
 
+#include "openvino/core/any.hpp"
 #include "openvino/core/node.hpp"
 #include "openvino/frontend/decoder.hpp"
 #include "openvino/frontend/gguf/visibility.hpp"
@@ -47,17 +48,17 @@ struct RopeConfig {
 // Following the established OpenVINO frontend pattern (cf. the PyTorch TorchDecoder + InputModel),
 // the translators see a GgufDecoder as a NODE decoder: visit_subgraph hands the visitor a fresh
 // decoder bound to a single node, and every per-node accessor (get_attribute, get_input_*,
-// get_output_*, get_op_*) refers to that node -- no node index is threaded through. The
-// MODEL-level questions (the graph's Parameter inputs, its output names, the shared RoPE config,
-// and node iteration) are asked through ov::frontend::gguf::InputModel, not by treating a decoder
-// instance as a "model decoder". The InputModel forwards those to the model-scope accessors below;
-// a concrete decoder answers them when queried before visit_subgraph binds it to a node.
+// get_output_*) refers to that node -- no node index is threaded through. The MODEL-level
+// questions (the graph's Parameter inputs, its output names, the shared RoPE config, and node
+// iteration) are asked through ov::frontend::gguf::InputModel, not by treating a decoder instance
+// as a "model decoder". The InputModel forwards those to the model-scope accessors below; a
+// concrete decoder answers them when queried before visit_subgraph binds it to a node.
 //
 // This is a typed, ggml-free interface: operation parameters are exposed through
 // get_attribute(name) / get_input_view_element_offset / get_output_shape / RopeConfig rather than
-// raw ggml `op_params` int32 arrays. A concrete decoder (e.g. the llama.cpp cgraph decoder) only
-// has to translate ggml's layout into these typed accessors -- the op translators never touch
-// ggml memory.
+// raw ggml `op_params` int32 arrays. A concrete decoder (e.g. the llama.cpp cgraph decoder, or the
+// native .gguf builder decoder) only has to translate ggml's layout into these typed accessors --
+// the op translators never touch ggml memory.
 class GGUF_FRONTEND_API GgufDecoder : public DecoderBase {
 public:
     // ── Node scope (the bound node; used by the op translators) ──────────────────────────────
@@ -104,26 +105,54 @@ class GGUF_FRONTEND_API GgufDecoder : public DecoderBase {
     // each node. This is the bridge from model scope to node scope.
     virtual void visit_subgraph(std::function<void(std::shared_ptr<GgufDecoder>)> node_visitor) const = 0;
 
-    // All model-scope input nodes: both primary inputs (Parameters) and auxiliary inputs
-    // (position IDs, KV-cache lengths, masks, etc.). Parameters are distinguished from auxiliary
-    // nodes by the caller via dynamic_pointer_cast<ov::op::v0::Parameter>.
+    // All model-scope primary input nodes (Parameters): the token/embedding input plus, on the
+    // stateful path, the KV-cache Parameters. Distinguished from auxiliary nodes by the caller via
+    // dynamic_pointer_cast<ov::op::v0::Parameter>.
     virtual const std::map<std::string, std::shared_ptr<ov::Node>>& get_model_inputs() const = 0;
+
     virtual std::vector<std::string> get_model_output_names() const = 0;
 
-    // NOTE: there is no get_model_weights(). A GGUF weight is surfaced as a regular node in
-    // visit_subgraph with the genuine ggml leaf op type "GGML_OP_NONE": the decoder marks it as a
-    // weight by exposing the raw weight bytes via get_attribute<ov::Tensor>("data"), the ggml
-    // quant type name via get_attribute<std::string>("quant_type") (e.g. "Q4_K", "F16") and the
-    // logical [rows, cols] shape via get_output_shape(). The frontend's translate_weight does the
-    // dequant / repacking / requantization, so the decoder never builds OV nodes itself. (Model
-    // inputs are also GGML_OP_NONE leaves, but they are returned via get_model_inputs() and
-    // resolved to Parameters before the walk, so they carry no "data".)
+    // ── Optional model scope ───────────────────────────────────────────────────────────────────
+    //
+    // The accessors below are how a decoder OPTIONALLY enriches the graph; each has a
+    // do-nothing default so a decoder only implements what it actually knows. That is what lets
+    // two very different decoders satisfy one interface: the native .gguf builder answers all of
+    // them, while the llama.cpp cgraph decoder (which is handed an already-built ggml graph and no
+    // GGUF metadata) answers none and is not forced to write empty stubs.
+    //
+    // Note what is NOT here: nothing describes the execution mode. There is no is_stateful /
+    // is_static, because a decoder describes ggml OPERATIONS, not a deployment. Conversion always
+    // yields a stateless graph; a caller that wants an OpenVINO KV cache registers
+    // ov::frontend::gguf::pass::MakeStateful as a DecoderTransformationExtension.
+
+    // Auxiliary model-scope inputs (position IDs, KV-cache lengths, attention masks). A decoder that
+    // folds these into get_model_inputs() leaves this empty. Note that beam_idx is not among them:
+    // it is a beam-search index into an OpenVINO state, which ggml has no counterpart for, so
+    // MakeStateful creates it rather than any decoder declaring it.
+    virtual const std::map<std::string, std::shared_ptr<ov::Node>>& get_model_extra_inputs() const {
+        return empty_node_map();
+    }
 
     // RoPE configuration, exposed through get_attribute<RopeConfig>("rope_config"):
     //   - at model scope (via InputModel::get_rope_config), used by TranslateSession::preprocess
     //     to pre-build the shared rope sin/cos table (skipped when RopeConfig::n_dims == 0, i.e.
     //     no RoPE, or per_op == true);
     //   - at node scope, the ROPE translator reads the same key for the op's own config.
+    //
+    // NOTE: weights are surfaced as GGML_OP_NONE leaves, by every decoder -- there is no separate
+    // weight accessor. A decoder marks such a leaf either with the raw ggml bytes
+    // (get_attribute<ov::Tensor>("data") + get_attribute<std::string>("quant_type") +
+    // get_output_shape(), the llama.cpp cgraph path) or with already-extracted weight/scales/zp
+    // tensors (get_attribute<bool>("gguf_weight") + "gguf.blob.<sub>" + "gguf_qtype", the native
+    // .gguf builder path). translate_weight accepts both payloads and builds the same compressed
+    // decompression subgraph from either.
+
+protected:
+    // Shared empty map backing the optional accessors above, which return by const reference.
+    static const std::map<std::string, std::shared_ptr<ov::Node>>& empty_node_map() {
+        static const std::map<std::string, std::shared_ptr<ov::Node>> empty;
+        return empty;
+    }
 };
 
 }  // namespace ov::frontend::gguf

src/frontends/gguf/include/openvino/frontend/gguf/frontend.hpp

@@ -34,6 +34,10 @@ class GGUF_FRONTEND_API FrontEnd : public ov::frontend::FrontEnd {
     /// - `ov::frontend::ConversionExtension` — registers a custom op translator for the
     ///   ggml op name given by `get_op_type()`.  The converter receives an
     ///   `ov::frontend::gguf::NodeContext` and returns an `ov::OutputVector`.
+    /// - `ov::frontend::DecoderTransformationExtension` — registers a normalization pass, run
+    ///   AHEAD of the frontend's built-in lowerings. This is how the execution mode is chosen: the
+    ///   frontend always converts to a stateless graph, and a caller that wants an OpenVINO KV
+    ///   cache registers `ov::frontend::gguf::pass::MakeStateful` (or its own variant) here.
     /// - `ov::frontend::TelemetryExtension` — receives error / event callbacks.
     /// - `ov::detail::SOExtension` — shared-library extension; its inner extension is
     ///   recursively registered.
@@ -44,17 +48,17 @@ class GGUF_FRONTEND_API FrontEnd : public ov::frontend::FrontEnd {
     void add_extension(const std::shared_ptr<ov::Extension>& extension) override;
 
 protected:
-    /// \brief Check if FrontEnd can recognize model from given parts.
-    /// \note Always returns false: this frontend is hidden from FrontEndManager and is never
-    ///       auto-selected. It is used only via direct linkage, by constructing FrontEnd and
-    ///       calling convert() on an InputModel built from a GgufDecoder.
-    /// \param variants Unused.
-    /// \return Always false.
+    /// \brief Check if FrontEnd can recognize the model from the given parts.
+    /// \param variants A single element holding a `std::shared_ptr<GgufDecoder>`. No other variant
+    ///        is recognized in this frontend: file-path (`.gguf`) loading is not yet implemented.
+    /// \return True iff variants holds exactly that; false otherwise.
     bool supported_impl(const std::vector<ov::Any>& variants) const override;
 
     /// \brief Load the input model from a GgufDecoder.
-    /// \param variants A single GgufDecoder (a .gguf file path is not accepted; the caller supplies
-    ///        the decoder). variants[0] must hold a std::shared_ptr<GgufDecoder>.
+    /// \param variants A single element holding a `std::shared_ptr<GgufDecoder>` -- a decoder
+    ///        supplied by a direct linker, wrapping an already-built ggml graph (the llama.cpp
+    ///        cgraph path). File-path (`.gguf`) loading, built per-architecture by the native
+    ///        builder, is not yet implemented in this frontend.
     /// \return InputModel::Ptr
     InputModel::Ptr load_impl(const std::vector<ov::Any>& variants) const override;
 

src/frontends/gguf/include/openvino/frontend/gguf/make_stateful.hpp

@@ -0,0 +1,50 @@
+// Copyright (C) 2018-2026 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#pragma once
+
+#include <memory>
+#include <set>
+#include <string>
+
+#include "openvino/frontend/gguf/visibility.hpp"
+#include "openvino/pass/pass.hpp"
+
+namespace ov::frontend::gguf::pass {
+
+
+
+class GGUF_FRONTEND_API MakeStateful : public ov::pass::ModelPass {
+public:
+    OPENVINO_MODEL_PASS_RTTI("gguf::MakeStateful");
+
+    /// \param skip_caches Friendly names of cache Parameters to leave stateless. A sliding-window
+    ///        cache is evicted from the front rather than only appended to, so an append-grown
+    ///        Variable would not reproduce it; such caches keep the stateless form.
+    /// \param append_axis Cache axis the new rows are appended along (the token axis). -1 infers it
+    ///        as the cache Parameter's single dynamic axis, which is how a graph that does not
+    ///        preallocate the cache states its token axis. Pass an explicit axis for a fully static
+    ///        (preallocated) cache, where there is nothing to infer from.
+    /// \param beam_idx_name Name of the beam-reorder input, which this pass ADDS to the model (it
+    ///        belongs to the state, so no decoder declares it; see the note above). The past cache is
+    ///        gathered by it along the batch axis before the append Concat. With batch 1 /
+    ///        beam_idx [0] that Gather is an identity, but emitting it is what lets CPU's
+    ///        stateful_sdpa_fusion match, and it is what makes beam search work. A model that
+    ///        already carries a Parameter of this name has it reused instead.
+    explicit MakeStateful(std::set<std::string> skip_caches = {},
+                          int64_t append_axis = -1,
+                          std::string beam_idx_name = "beam_idx")
+        : m_skip_caches(std::move(skip_caches)),
+          m_append_axis(append_axis),
+          m_beam_idx_name(std::move(beam_idx_name)) {}
+
+    bool run_on_model(const std::shared_ptr<ov::Model>& model) override;
+
+private:
+    std::set<std::string> m_skip_caches;
+    int64_t m_append_axis;
+    std::string m_beam_idx_name;
+};
+
+}  // namespace ov::frontend::gguf::pass

src/frontends/gguf/src/CMakeLists.txt

@@ -2,10 +2,9 @@
 # SPDX-License-Identifier: Apache-2.0
 #
 
-# LINKABLE_FRONTEND: installed alongside the other frontends so callers (the llama.cpp
-# ggml-openvino backend, OpenVINO GenAI) can link openvino::frontend::gguf directly. It is kept
-# out of the generic FrontEndManager loading API (FrontEndManager treats "gguf" as hidden); see
-# the discoverability note in frontend.cpp.
+# The GGUF frontend converts GGUF models to OpenVINO. It installs the library + headers alongside
+# the other frontends (LINKABLE_FRONTEND, no SKIP_INSTALL) so a direct linker -- the llama.cpp
+# ggml-openvino backend -- can link openvino::frontend::gguf and feed it a live GgufDecoder.
 ov_add_frontend(NAME gguf
                 LINKABLE_FRONTEND
                 FILEDESCRIPTION "FrontEnd to convert GGUF models"

src/frontends/gguf/src/frontend.cpp

@@ -4,6 +4,7 @@
 
 #include "openvino/frontend/gguf/frontend.hpp"
 
+
 #include "input_model.hpp"
 #include "op_table.hpp"
 #include "openvino/core/so_extension.hpp"
@@ -18,12 +19,17 @@ namespace ov {
 namespace frontend {
 namespace gguf {
 
-// Discoverability (intentional): consumed only by direct linkage -- a caller (the llama.cpp
-// ggml-openvino backend, OpenVINO GenAI) links openvino::frontend::gguf and feeds FrontEnd a live
-// GgufDecoder (no .gguf-path reader; see supported_impl / load_impl). It exports the standard
-// plugin entry points so FrontEndManager can scan the frontend dir without error, but "gguf" is
-// treated as hidden there (manager.cpp is_hidden_frontend), so it is never listed or auto-selected.
-// Drop it from that list once this frontend gains file-based loading and passes production review.
+// The frontend converts a live GgufDecoder -- supplied by a direct linker, currently the llama.cpp
+// ggml-openvino backend -- through the op translators below.
+//
+// Discoverability: "gguf" is in manager.cpp's is_hidden_frontend list, so it is not advertised by
+// available_front_ends() and not auto-selected by load_by_model -- core.read_model(".gguf") does
+// not resolve to it. It is still reachable explicitly, either by direct linkage (what GenAI and
+// the llama.cpp backend do) or by name via load_by_framework("gguf"). supported_impl below stays
+// implemented, so enabling core.read_model later is just dropping the name from that list.
+//
+// Driving the frontend directly needs no follow-up pass: normalization runs inside convert(), and
+// the only step read_model adds, update_v10_model(), fires solely for legacy IR v10.
 
 struct FrontEnd::Impl {
     std::unordered_map<std::string, CreatorFunction> op_extension_translators;
@@ -91,17 +97,14 @@ void FrontEnd::add_extension(const std::shared_ptr<ov::Extension>& extension) {
     }
 }
 
-bool FrontEnd::supported_impl(const std::vector<ov::Any>&) const {
-    // Always false: this frontend is never selected by FrontEndManager (load_by_model). It is used
-    // only via direct linkage -- a caller constructs FrontEnd and calls convert() with an
-    // InputModel built from a GgufDecoder -- which does not go through supported(). See the
-    // discoverability note at the top of this file.
-    return false;
+bool FrontEnd::supported_impl(const std::vector<ov::Any>& variants) const {
+    return !variants.empty() && variants[0].is<std::shared_ptr<GgufDecoder>>();
 }
 
 InputModel::Ptr FrontEnd::load_impl(const std::vector<ov::Any>& variants) const {
     FRONT_END_GENERAL_CHECK(!variants.empty(),
                             "GGUF Frontend requires at least one parameter in model representation.");
+
     FRONT_END_GENERAL_CHECK(variants[0].is<std::shared_ptr<GgufDecoder>>(),
                             "GGUF Frontend supports loading from a GgufDecoder only.");
     auto decoder = variants[0].as<std::shared_ptr<GgufDecoder>>();
@@ -113,10 +116,8 @@ InputModel::Ptr FrontEnd::load_impl(const std::vector<ov::Any>& variants) const
 }  // namespace frontend
 }  // namespace ov
 
-// Plugin registration. The frontend is installed in the frontend directory, so it must export
-// these entry points or FrontEndManager throws while scanning it. It registers as hidden (see the
-// discoverability note above): FrontEndManager loads it without error but never lists or
-// auto-selects it; only direct linkers use it.
+// Plugin registration. Exports the standard entry points so FrontEndManager can load the library;
+// selection is covered by the discoverability note at the top of this file.
 GGUF_FRONTEND_C_API ov::frontend::FrontEndVersion get_api_version() {
     return OV_FRONTEND_API_VERSION;
 }

src/frontends/gguf/src/input_model.cpp

@@ -34,6 +34,10 @@ void InputModel::visit_subgraph(const std::function<void(std::shared_ptr<GgufDec
     m_decoder->visit_subgraph(node_visitor);
 }
 
+const std::shared_ptr<GgufDecoder>& InputModel::get_model_decoder() const {
+    return m_decoder;
+}
+
 }  // namespace gguf
 }  // namespace frontend
 }  // namespace ov

src/frontends/gguf/src/input_model.hpp

@@ -7,12 +7,12 @@
 #include <functional>
 #include <map>
 #include <memory>
-#include "openvino/frontend/input_model.hpp"
 #include <string>
 #include <vector>
 
 #include "openvino/frontend/gguf/decoder.hpp"
 #include "openvino/frontend/gguf/visibility.hpp"
+#include "openvino/frontend/input_model.hpp"
 
 namespace ov::frontend::gguf {
 
@@ -36,6 +36,11 @@ class GGUF_FRONTEND_API InputModel : public ov::frontend::InputModel {
     RopeConfig get_rope_config() const;
     void visit_subgraph(const std::function<void(std::shared_ptr<GgufDecoder>)>& node_visitor) const;
 
+    // The underlying node-scoped decoder. TranslateSession uses it for the remaining model-scope
+    // questions that are only relevant on the native .gguf builder / stateful path (weights,
+    // extra inputs, KV param/result pairs, is_stateful / is_static, tokenizer metadata).
+    const std::shared_ptr<GgufDecoder>& get_model_decoder() const;
+
 private:
     std::shared_ptr<GgufDecoder> m_decoder;
 };