Dry sampler

.github/workflows/bench.yml

@@ -32,7 +32,7 @@ on:
     -  cron: '04 2 * * *'
 
 concurrency:
-  group: ${{ github.workflow }}-${{ github.head_ref && github.ref || github.run_id }}-${{ github.event.inputs.sha }}
+  group: ${{ github.workflow }}-${{ github.ref || github.run_id }}-${{ github.event.inputs.sha }}
   cancel-in-progress: true
 
 jobs:

.github/workflows/build.yml

@@ -32,6 +32,8 @@ jobs:
       - name: Clone
         id: checkout
         uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
 
       - name: Dependencies
         id: depends
@@ -88,6 +90,8 @@ jobs:
       - name: Clone
         id: checkout
         uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
 
       - name: Dependencies
         id: depends
@@ -206,6 +210,8 @@ jobs:
       - name: Clone
         id: checkout
         uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
 
       - name: Dependencies
         id: depends
@@ -238,6 +244,33 @@ jobs:
           ./bin/convert-llama2c-to-ggml --copy-vocab-from-model ./tok512.bin --llama2c-model stories260K.bin --llama2c-output-model stories260K.gguf
           ./bin/main -m stories260K.gguf -p "One day, Lily met a Shoggoth" -n 500 -c 256
 
+      - name: Determine tag name
+        id: tag
+        shell: bash
+        run: |
+          BUILD_NUMBER="$(git rev-list --count HEAD)"
+          SHORT_HASH="$(git rev-parse --short=7 HEAD)"
+          if [[ "${{ env.BRANCH_NAME }}" == "master" ]]; then
+            echo "name=b${BUILD_NUMBER}" >> $GITHUB_OUTPUT
+          else
+            SAFE_NAME=$(echo "${{ env.BRANCH_NAME }}" | tr '/' '-')
+            echo "name=${SAFE_NAME}-b${BUILD_NUMBER}-${SHORT_HASH}" >> $GITHUB_OUTPUT
+          fi
+
+      - name: Pack artifacts
+        id: pack_artifacts
+        if: ${{ ( github.event_name == 'push' && github.ref == 'refs/heads/master' ) || github.event.inputs.create_release == 'true' }}
+        run: |
+          cp LICENSE ./build/bin/
+          zip -r llama-${{ steps.tag.outputs.name }}-bin-ubuntu-x64.zip ./build/bin/*
+
+      - name: Upload artifacts
+        if: ${{ ( github.event_name == 'push' && github.ref == 'refs/heads/master' ) || github.event.inputs.create_release == 'true' }}
+        uses: actions/upload-artifact@v4
+        with:
+          path: llama-${{ steps.tag.outputs.name }}-bin-ubuntu-x64.zip
+          name: llama-bin-ubuntu-x64.zip
+
 #  ubuntu-latest-cmake-sanitizer:
 #    runs-on: ubuntu-latest
 #

.github/workflows/server.yml

@@ -23,7 +23,7 @@ on:
     -  cron: '2 4 * * *'
 
 concurrency:
-  group: ${{ github.workflow }}-${{ github.head_ref && github.ref || github.run_id }}
+  group: ${{ github.workflow }}-${{ github.ref || github.run_id }}
   cancel-in-progress: true
 
 jobs:

.gitignore

@@ -34,6 +34,7 @@ lcov-report/
 gcovr-report/
 
 build*
+!build.zig
 cmake-build-*
 out/
 tmp/
@@ -101,6 +102,9 @@ qnt-*.txt
 perf-*.txt
 
 examples/jeopardy/results.txt
+examples/server/*.html.hpp
+examples/server/*.js.hpp
+examples/server/*.mjs.hpp
 
 poetry.lock
 poetry.toml

CMakeLists.txt

@@ -43,17 +43,11 @@ else()
     set(LLAMA_METAL_DEFAULT OFF)
 endif()
 
-# TODO: fix this for Android CI
-#       https://github.com/ggerganov/llama.cpp/pull/6716#issuecomment-2061509191
-#if (CMAKE_SYSTEM_NAME MATCHES "ANDROID")
-#    set(LLAMA_LLAMAFILE_DEFAULT OFF)
-#else()
-#    set(LLAMA_LLAMAFILE_DEFAULT ON)
-#endif()
-
-# TODO: temporary disable until MoE is fixed
-#       https://github.com/ggerganov/llama.cpp/pull/6716
-set(LLAMA_LLAMAFILE_DEFAULT OFF)
+if (CMAKE_SYSTEM_NAME MATCHES "ANDROID")
+    set(LLAMA_LLAMAFILE_DEFAULT OFF)
+else()
+    set(LLAMA_LLAMAFILE_DEFAULT ON)
+endif()
 
 # general
 option(BUILD_SHARED_LIBS                "build shared libraries"                                OFF)

Makefile

@@ -384,10 +384,6 @@ ifdef LLAMA_OPENBLAS
 	MK_LDFLAGS  += $(shell pkg-config --libs openblas)
 endif # LLAMA_OPENBLAS
 
-# TODO: temporary disable until MoE is fixed
-#       https://github.com/ggerganov/llama.cpp/pull/6716
-LLAMA_NO_LLAMAFILE := 1
-
 ifndef LLAMA_NO_LLAMAFILE
 	MK_CPPFLAGS += -DGGML_USE_LLAMAFILE
 	OBJS        += sgemm.o
@@ -699,7 +695,7 @@ OBJS += ggml-alloc.o ggml-backend.o ggml-quants.o unicode.o unicode-data.o
 llama.o: llama.cpp unicode.h ggml.h ggml-alloc.h ggml-backend.h ggml-cuda.h ggml-metal.h llama.h
 	$(CXX) $(CXXFLAGS) -c $< -o $@
 
-COMMON_H_DEPS = common/common.h common/sampling.h common/log.h
+COMMON_H_DEPS = common/common.h common/sampling.h common/log.h llama.h
 COMMON_DEPS   = common.o sampling.o grammar-parser.o build-info.o json-schema-to-grammar.o
 
 common.o: common/common.cpp $(COMMON_H_DEPS)
@@ -800,10 +796,19 @@ save-load-state: examples/save-load-state/save-load-state.cpp ggml.o llama.o $(C
 	$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)
 
-server: examples/server/server.cpp examples/server/utils.hpp examples/server/httplib.h common/json.hpp examples/server/index.html.hpp examples/server/index.js.hpp examples/server/completion.js.hpp common/stb_image.h ggml.o llama.o $(COMMON_DEPS) grammar-parser.o $(OBJS)
+server: examples/server/server.cpp examples/server/utils.hpp examples/server/httplib.h common/json.hpp examples/server/index.html.hpp examples/server/index.js.hpp examples/server/completion.js.hpp examples/server/json-schema-to-grammar.mjs.hpp common/stb_image.h ggml.o llama.o $(COMMON_DEPS) grammar-parser.o $(OBJS)
 	$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h %.hpp $<,$^) -Iexamples/server $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS) $(LWINSOCK2)
 
+# Portable equivalent of `cd examples/server/public && xxd -i $(notdir $<) ../$(notdir $<).hpp`:
+examples/server/%.hpp: examples/server/public/% Makefile
+	@( export NAME=$(subst .,_,$(subst -,_,$(notdir $<))) && \
+		echo "unsigned char $${NAME}[] = {" && \
+		cat $< | od -v -t x1 -An | sed -E 's/([0-9a-fA-F]+)/0x\1, /g' && \
+		echo "};" && \
+		echo "unsigned int $${NAME}_len = $(shell cat $< | wc -c );" \
+	) > $@
+
 gguf: examples/gguf/gguf.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) -c $< -o $(call GET_OBJ_FILE, $<)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h $<,$^) $(call GET_OBJ_FILE, $<) -o $@ $(LDFLAGS)

README-sycl.md

@@ -229,12 +229,11 @@ source /opt/intel/oneapi/setvars.sh
 # Build LLAMA with MKL BLAS acceleration for intel GPU
 mkdir -p build && cd build
 
-# Option 1: Use FP16 for better performance in long-prompt  inference
-cmake --build .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
-# Or without "--build", run "make" next
+# Option 1: Use FP32 (recommended for better performance in most cases)
+cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 
-# Option 2: Use FP32 by default
-cmake --build .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
+# Option 2: Use FP16
+cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
 
 #build all binary
 cmake --build . --config Release -j -v
@@ -251,11 +250,11 @@ export CPLUS_INCLUDE_DIR=/path/to/oneMKL/include:$CPLUS_INCLUDE_DIR
 # Build LLAMA with Nvidia BLAS acceleration through SYCL
 mkdir -p build && cd build
 
-# Option 1: Use FP16 for better performance in long-prompt  inference
-cmake --build .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
+# Option 1: Use FP32 (recommended for better performance in most cases)
+cmake .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
 
-# Option 2: Use FP32 by default
-cmake --build .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
+# Option 2: Use FP16
+cmake .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
 
 #build all binary
 cmake --build . --config Release -j -v
@@ -417,6 +416,10 @@ mkdir -p build
 cd build
 @call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" intel64 --force
 
+# Option 1: Use FP32 (recommended for better performance in most cases)
+cmake -G "MinGW Makefiles" ..  -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release
+
+# Option 2: Or FP16
 cmake -G "MinGW Makefiles" ..  -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release -DLLAMA_SYCL_F16=ON
 
 make -j

README.md

@@ -10,6 +10,7 @@ Inference of Meta's [LLaMA](https://arxiv.org/abs/2302.13971) model (and others)
 
 ### Recent API changes
 
+- [2024 Apr 21] `llama_token_to_piece` can now optionally render special tokens https://github.com/ggerganov/llama.cpp/pull/6807
 - [2024 Apr 4] State and session file functions reorganized under `llama_state_*` https://github.com/ggerganov/llama.cpp/pull/6341
 - [2024 Mar 26] Logits and embeddings API updated for compactness https://github.com/ggerganov/llama.cpp/pull/6122
 - [2024 Mar 13] Add `llama_synchronize()` + `llama_context_params.n_ubatch` https://github.com/ggerganov/llama.cpp/pull/6017
@@ -95,7 +96,7 @@ Typically finetunes of the base models below are supported as well.
 - [X] [Mistral 7B](https://huggingface.co/mistralai/Mistral-7B-v0.1)
 - [x] [Mixtral MoE](https://huggingface.co/models?search=mistral-ai/Mixtral)
 - [x] [DBRX](https://huggingface.co/databricks/dbrx-instruct)
-- [X] Falcon
+- [X] [Falcon](https://huggingface.co/models?search=tiiuae/falcon)
 - [X] [Chinese LLaMA / Alpaca](https://github.com/ymcui/Chinese-LLaMA-Alpaca) and [Chinese LLaMA-2 / Alpaca-2](https://github.com/ymcui/Chinese-LLaMA-Alpaca-2)
 - [X] [Vigogne (French)](https://github.com/bofenghuang/vigogne)
 - [X] [Koala](https://bair.berkeley.edu/blog/2023/04/03/koala/)
@@ -122,6 +123,7 @@ Typically finetunes of the base models below are supported as well.
 - [x] [Command-R](https://huggingface.co/CohereForAI/c4ai-command-r-v01)
 - [x] [SEA-LION](https://huggingface.co/models?search=sea-lion)
 - [x] [GritLM-7B](https://huggingface.co/GritLM/GritLM-7B) + [GritLM-8x7B](https://huggingface.co/GritLM/GritLM-8x7B)
+- [x] [OLMo](https://allenai.org/olmo)
 
 (instructions for supporting more models: [HOWTO-add-model.md](./docs/HOWTO-add-model.md))
 
@@ -548,7 +550,7 @@ Building the program with BLAS support may lead to some performance improvements
   OpenCL acceleration is provided by the matrix multiplication kernels from the [CLBlast](https://github.com/CNugteren/CLBlast) project and custom kernels for ggml that can generate tokens on the GPU.
 
   You will need the [OpenCL SDK](https://github.com/KhronosGroup/OpenCL-SDK).
-    - For Ubuntu or Debian, the packages `opencl-headers`, `ocl-icd` may be needed.
+    - For Ubuntu, Debian, and Fedora the packages `opencl-headers`, `ocl-icd` may be needed.
 
     - For Windows, a pre-built SDK is available on the [OpenCL Releases](https://github.com/KhronosGroup/OpenCL-SDK/releases) page.
 
@@ -573,6 +575,12 @@ Building the program with BLAS support may lead to some performance improvements
 
   Pre-built CLBlast binaries may be found on the [CLBlast Releases](https://github.com/CNugteren/CLBlast/releases) page. For Unix variants, it may also be found in your operating system's packages.
 
+  Linux packaging:
+  Fedora Linux:
+  ```bash
+  sudo dnf install clblast
+  ```
+
   Alternatively, they may be built from source.
 
   - <details>

build.zig

@@ -140,4 +140,33 @@ pub fn build(b: *std.build.Builder) !void {
     if (server.target.isWindows()) {
         server.linkSystemLibrary("ws2_32");
     }
+
+    const server_assets = [_][]const u8{ "index.html", "index.js", "completion.js", "json-schema-to-grammar.mjs" };
+    for (server_assets) |asset| {
+        const input_path = b.fmt("examples/server/public/{s}", .{asset});
+        const output_path = b.fmt("examples/server/{s}.hpp", .{asset});
+
+        // Portable equivalent of `b.addSystemCommand(&.{ "xxd", "-n", asset, "-i", input_path, output_path }) })`:
+
+        const input = try std.fs.cwd().readFileAlloc(b.allocator, input_path, std.math.maxInt(usize));
+        defer b.allocator.free(input);
+
+        var buf = std.ArrayList(u8).init(b.allocator);
+        defer buf.deinit();
+
+        for (input) |byte| {
+            try std.fmt.format(buf.writer(), "0x{X:0>2}, ", .{byte});
+        }
+
+        var name = try std.mem.replaceOwned(u8, b.allocator, asset, "-", "_");
+        defer b.allocator.free(name);
+        std.mem.replaceScalar(u8, name, '.', '_');
+
+        try std.fs.cwd().writeFile(output_path, b.fmt(
+            "unsigned char {s}[] = {{{s}}};\nunsigned int {s}_len = {d};\n",
+            .{ name, buf.items, name, input.len },
+        ));
+
+        std.debug.print("Dumped hex of \"{s}\" ({s}) to {s}\n", .{ input_path, name, output_path });
+    }
 }

common/common.cpp

@@ -108,7 +108,7 @@ int32_t get_num_physical_cores() {
     return n_threads > 0 ? (n_threads <= 4 ? n_threads : n_threads / 2) : 4;
 }
 
-#if defined(__x86_64__) && defined(__linux__)
+#if defined(__x86_64__) && defined(__linux__) && !defined(__ANDROID__)
 #include <pthread.h>
 
 static void cpuid(unsigned leaf, unsigned subleaf,
@@ -162,7 +162,7 @@ static int count_math_cpus(int cpu_count) {
  * Returns number of CPUs on system that are useful for math.
  */
 int get_math_cpu_count() {
-#if defined(__x86_64__) && defined(__linux__)
+#if defined(__x86_64__) && defined(__linux__) && !defined(__ANDROID__)
     int cpu_count = sysconf(_SC_NPROCESSORS_ONLN);
     if (cpu_count < 1) {
         return get_num_physical_cores();
@@ -582,6 +582,30 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
         sparams.penalty_present = std::stof(argv[i]);
         return true;
     }
+    if (arg == "--dry-multiplier") {
+        if (++i >= argc) {
+            invalid_param = true;
+            return true;
+        }
+        sparams.dry_multiplier = std::stof(argv[i]);
+        return true;
+    }
+    if (arg == "--dry-base") {
+        if (++i >= argc) {
+            invalid_param = true;
+            return true;
+        }
+        sparams.dry_base = std::stoi(argv[i]);
+        return true;
+    }
+    if (arg == "--dry-allowed-length") {
+        if (++i >= argc) {
+            invalid_param = true;
+            return true;
+        }
+        sparams.dry_allowed_length = std::stoi(argv[i]);
+        return true;
+    }
     if (arg == "--dynatemp-range") {
         if (++i >= argc) {
             invalid_param = true;
@@ -1429,6 +1453,10 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     printf("  --repeat-penalty N    penalize repeat sequence of tokens (default: %.1f, 1.0 = disabled)\n", (double)sparams.penalty_repeat);
     printf("  --presence-penalty N  repeat alpha presence penalty (default: %.1f, 0.0 = disabled)\n", (double)sparams.penalty_present);
     printf("  --frequency-penalty N repeat alpha frequency penalty (default: %.1f, 0.0 = disabled)\n", (double)sparams.penalty_freq);
+    printf("  --dry-multiplier N    DRY sampler multiplier (default: %.1f, 0.0 = disabled)\n", (double)sparams.dry_multiplier);
+    printf("  --dry-base N          DRY sampler base (default: %.1f)\n", (double)sparams.dry_base);
+    printf("  --dry-allowed-length N\n");
+    printf("                        DRY sampler allowed length (default: %d)\n", sparams.dry_allowed_length);
     printf("  --dynatemp-range N    dynamic temperature range (default: %.1f, 0.0 = disabled)\n", (double)sparams.dynatemp_range);
     printf("  --dynatemp-exp N      dynamic temperature exponent (default: %.1f)\n", (double)sparams.dynatemp_exponent);
     printf("  --mirostat N          use Mirostat sampling.\n");
@@ -2333,10 +2361,10 @@ std::vector<llama_token> llama_tokenize(
 
 std::string llama_token_to_piece(const struct llama_context * ctx, llama_token token) {
     std::vector<char> result(8, 0);
-    const int n_tokens = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size());
+    const int n_tokens = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size(), true);
     if (n_tokens < 0) {
         result.resize(-n_tokens);
-        int check = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size());
+        int check = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size(), true);
         GGML_ASSERT(check == -n_tokens);
     } else {
         result.resize(n_tokens);

common/sampling.cpp

@@ -103,10 +103,10 @@ std::string llama_sampling_print(const llama_sampling_params & params) {
     snprintf(result, sizeof(result),
             "\trepeat_last_n = %d, repeat_penalty = %.3f, frequency_penalty = %.3f, presence_penalty = %.3f\n"
             "\ttop_k = %d, tfs_z = %.3f, top_p = %.3f, min_p = %.3f, typical_p = %.3f, temp = %.3f\n"
-            "\tmirostat = %d, mirostat_lr = %.3f, mirostat_ent = %.3f",
+            "\tmirostat = %d, mirostat_lr = %.3f, mirostat_ent = %.3f, dry_multiplier = %.3f, dry_base = %.3f, dry_allowed_length = %d",
             params.penalty_last_n, params.penalty_repeat, params.penalty_freq, params.penalty_present,
             params.top_k, params.tfs_z, params.top_p, params.min_p, params.typical_p, params.temp,
-            params.mirostat, params.mirostat_eta, params.mirostat_tau);
+            params.mirostat, params.mirostat_eta, params.mirostat_tau, params.dry_multiplier, params.dry_base, params.dry_allowed_length);
 
     return std::string(result);
 }
@@ -314,12 +314,11 @@ static llama_token_data_array llama_sampling_prepare_impl(
 
         // DRY penalties (multiplier > 0 means enabled)
         if(dry_multiplier > 0.0f) {
-            llama_sample_dry(ctx_main, &cur_p,
+                llama_sample_dry(&cur_p,
                             penalty_tokens.data() + penalty_tokens.size() - penalty_tokens_used_size,
                             penalty_tokens_used_size, dry_base, dry_multiplier, dry_allowed_length,
                             params.dry_sequence_breakers.data(), params.dry_sequence_breakers.size());
         }
-
 
         if (!penalize_nl) {
             for (size_t idx = 0; idx < cur_p.size; idx++) {