talk-llama : sync llama.cpp

ggml-ci
2025-06-22 16:38:58 +00:00 · 2025-06-18 10:22:47 +03:00
parent 69061e356f
commit 2f60ebc3c2
26 changed files with 1454 additions and 504 deletions
--- a/examples/talk-llama/llama-quant.cpp
+++ b/examples/talk-llama/llama-quant.cpp
@ -585,7 +585,8 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
            if (o.tag == LLAMA_KV_OVERRIDE_TYPE_FLOAT) {
                gguf_set_val_f32(ctx_out.get(), o.key, o.val_f64);
            } else if (o.tag == LLAMA_KV_OVERRIDE_TYPE_INT) {
-                gguf_set_val_i32(ctx_out.get(), o.key, o.val_i64);
+                // Setting type to UINT32. See https://github.com/ggml-org/llama.cpp/pull/14182 for context
+                gguf_set_val_u32(ctx_out.get(), o.key, (uint32_t)abs(o.val_i64));
            } else if (o.tag == LLAMA_KV_OVERRIDE_TYPE_BOOL) {
                gguf_set_val_bool(ctx_out.get(), o.key, o.val_bool);
            } else if (o.tag == LLAMA_KV_OVERRIDE_TYPE_STR) {