@qvac/llm-llamacpp 0.61.1 → 0.61.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (18) hide show
  1. package/README.md +2 -0
  2. package/index.d.ts +11 -0
  3. package/package.json +1 -1
  4. package/prebuilds/android-arm64/qvac__llm-llamacpp.bare +0 -0
  5. package/prebuilds/darwin-arm64/qvac__llm-llamacpp.bare +0 -0
  6. package/prebuilds/darwin-arm64/qvac__llm-llamacpp.bare.exports +1 -2
  7. package/prebuilds/darwin-x64/qvac__llm-llamacpp.bare +0 -0
  8. package/prebuilds/darwin-x64/qvac__llm-llamacpp.bare.exports +1 -2
  9. package/prebuilds/ios-arm64/qvac__llm-llamacpp.bare +0 -0
  10. package/prebuilds/ios-arm64/qvac__llm-llamacpp.bare.exports +1 -2
  11. package/prebuilds/ios-arm64-simulator/qvac__llm-llamacpp.bare +0 -0
  12. package/prebuilds/ios-arm64-simulator/qvac__llm-llamacpp.bare.exports +1 -2
  13. package/prebuilds/ios-x64-simulator/qvac__llm-llamacpp.bare +0 -0
  14. package/prebuilds/ios-x64-simulator/qvac__llm-llamacpp.bare.exports +1 -2
  15. package/prebuilds/linux-arm64/qvac__llm-llamacpp.bare +0 -0
  16. package/prebuilds/linux-x64/qvac__llm-llamacpp.bare +0 -0
  17. package/prebuilds/win32-x64/qvac__llm-llamacpp.bare +0 -0
  18. package/prebuilds/win32-x64/qvac__llm-llamacpp.bare.exports +0 -0
package/README.md CHANGED
@@ -180,6 +180,8 @@ const config = {
180
180
  | device | `"gpu"` or `"cpu"` | — (required) | Device to run inference on |
181
181
  | gpu_layers | integer | 0 | Number of model layers to offload to GPU |
182
182
  | ctx_size | 0 – model-dependent | 4096 (0 = loaded from model) | Context window size |
183
+ | poll | integer 0 – 100 | 0 | Main GGML threadpool busy-wait level. `0` sleeps while idle; higher values can reduce wake latency but consume CPU during GPU work |
184
+ | poll-batch | integer 0 – 100 | `poll` | Batch GGML threadpool busy-wait level |
183
185
  | lora | string | — | Path to LoRA adapter file |
184
186
  | temp | 0.00 – 2.00 | 0.8 | Sampling temperature |
185
187
  | top_p | 0 – 1 | 0.9 | Top-p (nucleus) sampling |
package/index.d.ts CHANGED
@@ -141,6 +141,17 @@ declare namespace LlmLlamacpp {
141
141
  backend?: string;
142
142
  gpu_layers?: NumericLike;
143
143
  ctx_size?: NumericLike;
144
+ /**
145
+ * Main GGML threadpool busy-wait level, from 0 (sleep while idle) to 100
146
+ * (maximum polling). Defaults to 0. Values above 0 can reduce worker wake
147
+ * latency for CPU-heavy workloads, but consume CPU while GPU work runs.
148
+ */
149
+ poll?: NumericLike;
150
+ /**
151
+ * Batch GGML threadpool busy-wait level. Defaults to `poll`, or 0 when
152
+ * neither option is set.
153
+ */
154
+ "poll-batch"?: NumericLike;
144
155
  system_prompt?: string;
145
156
  lora?: string;
146
157
  temp?: NumericLike;
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@qvac/llm-llamacpp",
3
- "version": "0.61.1",
3
+ "version": "0.61.2",
4
4
  "description": "llama addon for qvac",
5
5
  "addon": true,
6
6
  "scripts": {
package/prebuilds/darwin-arm64/qvac__llm-llamacpp.bare.exports CHANGED
@@ -7,7 +7,7 @@
7
7
  ],
8
8
  "current_versions": [
9
9
  {
10
- "version": "0.61.1"
10
+ "version": "0.61.2"
11
11
  }
12
12
  ],
13
13
  "exported_symbols": [
@@ -84,7 +84,6 @@
84
84
  "__ZN24llama_finetuning_helpers23optEpochCallbackWrapperEbP16ggml_opt_contextP16ggml_opt_datasetP15ggml_opt_resultxxx",
85
85
  "__ZThn24_N10LlamaModelD0Ev",
86
86
  "__ZN14MtmdLlmContext8setNPastEi",
87
- "__ZN14TextLlmContext26initializeOwnedThreadpoolsEv",
88
87
  "__ZN30qvac_lib_inference_addon_llama8batching19MultiRequestBatcher10addRequestEONSt3__16vectorIiNS2_9allocatorIiEEEERj",
89
88
  "__ZN30qvac_lib_inference_addon_llama8batching19MultiRequestBatcher19sampleAndAppendIdleERKNSt3__18functionIFijiEEE",
90
89
  "__ZN10LlamaModel30discardFinetuneCancelSaveModesERKNSt3__16vectorIyNS0_9allocatorIyEEEE",
package/prebuilds/darwin-x64/qvac__llm-llamacpp.bare.exports CHANGED
@@ -7,7 +7,7 @@
7
7
  ],
8
8
  "current_versions": [
9
9
  {
10
- "version": "0.61.1"
10
+ "version": "0.61.2"
11
11
  }
12
12
  ],
13
13
  "exported_symbols": [
@@ -798,7 +798,6 @@
798
798
  "__ZN24llama_finetuning_helpers23optEpochCallbackWrapperEbP16ggml_opt_contextP16ggml_opt_datasetP15ggml_opt_resultxxx",
799
799
  "__ZN14MtmdLlmContext8setNPastEi",
800
800
  "__ZThn24_N10LlamaModelD0Ev",
801
- "__ZN14TextLlmContext26initializeOwnedThreadpoolsEv",
802
801
  "__ZN30qvac_lib_inference_addon_llama8batching7RequestC1EjO11PrefillPlanji",
803
802
  "__ZN30qvac_lib_inference_addon_llama5utils25partialSnapshotFromStreamERNSt3__113basic_istreamIcNS1_11char_traitsIcEEEEyiNS0_15SnapshotStorageERKNS1_12basic_stringIcS4_NS1_9allocatorIcEEEERKNS1_8functionIFvPKhmEEERNS0_21SequenceStateSnapshotE",
804
803
  "__ZN10LlamaModelD0Ev",
package/prebuilds/ios-arm64/qvac__llm-llamacpp.bare.exports CHANGED
@@ -7,7 +7,7 @@
7
7
  ],
8
8
  "current_versions": [
9
9
  {
10
- "version": "0.61.1"
10
+ "version": "0.61.2"
11
11
  }
12
12
  ],
13
13
  "exported_symbols": [
@@ -84,7 +84,6 @@
84
84
  "__ZN24llama_finetuning_helpers23optEpochCallbackWrapperEbP16ggml_opt_contextP16ggml_opt_datasetP15ggml_opt_resultxxx",
85
85
  "__ZThn24_N10LlamaModelD0Ev",
86
86
  "__ZN14MtmdLlmContext8setNPastEi",
87
- "__ZN14TextLlmContext26initializeOwnedThreadpoolsEv",
88
87
  "__ZN30qvac_lib_inference_addon_llama8batching19MultiRequestBatcher10addRequestEONSt3__16vectorIiNS2_9allocatorIiEEEERj",
89
88
  "__ZN30qvac_lib_inference_addon_llama8batching19MultiRequestBatcher19sampleAndAppendIdleERKNSt3__18functionIFijiEEE",
90
89
  "__ZN10LlamaModel30discardFinetuneCancelSaveModesERKNSt3__16vectorIyNS0_9allocatorIyEEEE",
package/prebuilds/ios-arm64-simulator/qvac__llm-llamacpp.bare.exports CHANGED
@@ -7,7 +7,7 @@
7
7
  ],
8
8
  "current_versions": [
9
9
  {
10
- "version": "0.61.1"
10
+ "version": "0.61.2"
11
11
  }
12
12
  ],
13
13
  "exported_symbols": [
@@ -84,7 +84,6 @@
84
84
  "__ZN24llama_finetuning_helpers23optEpochCallbackWrapperEbP16ggml_opt_contextP16ggml_opt_datasetP15ggml_opt_resultxxx",
85
85
  "__ZThn24_N10LlamaModelD0Ev",
86
86
  "__ZN14MtmdLlmContext8setNPastEi",
87
- "__ZN14TextLlmContext26initializeOwnedThreadpoolsEv",
88
87
  "__ZN30qvac_lib_inference_addon_llama8batching19MultiRequestBatcher10addRequestEONSt3__16vectorIiNS2_9allocatorIiEEEERj",
89
88
  "__ZN30qvac_lib_inference_addon_llama8batching19MultiRequestBatcher19sampleAndAppendIdleERKNSt3__18functionIFijiEEE",
90
89
  "__ZN10LlamaModel30discardFinetuneCancelSaveModesERKNSt3__16vectorIyNS0_9allocatorIyEEEE",
package/prebuilds/ios-x64-simulator/qvac__llm-llamacpp.bare.exports CHANGED
@@ -7,7 +7,7 @@
7
7
  ],
8
8
  "current_versions": [
9
9
  {
10
- "version": "0.61.1"
10
+ "version": "0.61.2"
11
11
  }
12
12
  ],
13
13
  "exported_symbols": [
@@ -811,7 +811,6 @@
811
811
  "__ZN24llama_finetuning_helpers23optEpochCallbackWrapperEbP16ggml_opt_contextP16ggml_opt_datasetP15ggml_opt_resultxxx",
812
812
  "__ZN14MtmdLlmContext8setNPastEi",
813
813
  "__ZThn24_N10LlamaModelD0Ev",
814
- "__ZN14TextLlmContext26initializeOwnedThreadpoolsEv",
815
814
  "__ZN30qvac_lib_inference_addon_llama8batching7RequestC1EjO11PrefillPlanji",
816
815
  "__ZN30qvac_lib_inference_addon_llama5utils25partialSnapshotFromStreamERNSt3__113basic_istreamIcNS1_11char_traitsIcEEEEyiNS0_15SnapshotStorageERKNS1_12basic_stringIcS4_NS1_9allocatorIcEEEERKNS1_8functionIFvPKhmEEERNS0_21SequenceStateSnapshotE",
817
816
  "__ZN10LlamaModelD0Ev",