退回到 b7516 版本
This commit is contained in:
@@ -78,7 +78,6 @@ json task_params::to_json(bool only_metrics) const {
|
||||
{"speculative.p_min", speculative.p_min},
|
||||
{"timings_per_token", timings_per_token},
|
||||
{"post_sampling_probs", post_sampling_probs},
|
||||
{"backend_sampling", sampling.backend_sampling},
|
||||
{"lora", lora},
|
||||
};
|
||||
}
|
||||
@@ -137,7 +136,6 @@ json task_params::to_json(bool only_metrics) const {
|
||||
{"speculative.p_min", speculative.p_min},
|
||||
{"timings_per_token", timings_per_token},
|
||||
{"post_sampling_probs", post_sampling_probs},
|
||||
{"backend_sampling", sampling.backend_sampling},
|
||||
{"lora", lora},
|
||||
};
|
||||
}
|
||||
@@ -160,7 +158,6 @@ task_params server_task::params_from_json_cmpl(
|
||||
defaults.n_keep = params_base.n_keep;
|
||||
defaults.n_predict = params_base.n_predict;
|
||||
defaults.n_cache_reuse = params_base.n_cache_reuse;
|
||||
defaults.cache_prompt = params_base.cache_prompt;
|
||||
defaults.antiprompt = params_base.antiprompt;
|
||||
|
||||
// enabling this will output extra debug information in the HTTP responses from the server
|
||||
@@ -170,7 +167,7 @@ task_params server_task::params_from_json_cmpl(
|
||||
params.stream = json_value(data, "stream", false);
|
||||
auto stream_opt = json_value(data, "stream_options", json::object());
|
||||
params.include_usage = json_value(stream_opt, "include_usage", false);
|
||||
params.cache_prompt = json_value(data, "cache_prompt", defaults.cache_prompt);
|
||||
params.cache_prompt = json_value(data, "cache_prompt", true);
|
||||
params.return_tokens = json_value(data, "return_tokens", false);
|
||||
params.return_progress = json_value(data, "return_progress", false);
|
||||
params.n_predict = json_value(data, "n_predict", json_value(data, "max_tokens", defaults.n_predict));
|
||||
@@ -204,12 +201,9 @@ task_params server_task::params_from_json_cmpl(
|
||||
params.sampling.mirostat = json_value(data, "mirostat", defaults.sampling.mirostat);
|
||||
params.sampling.mirostat_tau = json_value(data, "mirostat_tau", defaults.sampling.mirostat_tau);
|
||||
params.sampling.mirostat_eta = json_value(data, "mirostat_eta", defaults.sampling.mirostat_eta);
|
||||
params.sampling.adaptive_target = json_value(data, "adaptive_target", defaults.sampling.adaptive_target);
|
||||
params.sampling.adaptive_decay = json_value(data, "adaptive_decay", defaults.sampling.adaptive_decay);
|
||||
params.sampling.seed = json_value(data, "seed", defaults.sampling.seed);
|
||||
params.sampling.n_probs = json_value(data, "n_probs", defaults.sampling.n_probs);
|
||||
params.sampling.min_keep = json_value(data, "min_keep", defaults.sampling.min_keep);
|
||||
params.sampling.backend_sampling = json_value(data, "backend_sampling", defaults.sampling.backend_sampling);
|
||||
params.post_sampling_probs = json_value(data, "post_sampling_probs", defaults.post_sampling_probs);
|
||||
|
||||
params.speculative.n_min = json_value(data, "speculative.n_min", defaults.speculative.n_min);
|
||||
@@ -817,15 +811,6 @@ json server_task_result_cmpl_final::to_json_anthropic() {
|
||||
msg.content = content;
|
||||
}
|
||||
|
||||
// thinking block comes first (Anthropic extended thinking format)
|
||||
if (!msg.reasoning_content.empty()) {
|
||||
content_blocks.push_back({
|
||||
{"type", "thinking"},
|
||||
{"thinking", msg.reasoning_content},
|
||||
{"signature", ""} // empty signature for local models (no cryptographic verification)
|
||||
});
|
||||
}
|
||||
|
||||
if (!msg.content.empty()) {
|
||||
content_blocks.push_back({
|
||||
{"type", "text"},
|
||||
@@ -874,57 +859,20 @@ json server_task_result_cmpl_final::to_json_anthropic_stream() {
|
||||
stop_reason = oaicompat_msg.tool_calls.empty() ? "end_turn" : "tool_use";
|
||||
}
|
||||
|
||||
bool has_thinking = !oaicompat_msg.reasoning_content.empty();
|
||||
bool has_text = !oaicompat_msg.content.empty();
|
||||
bool has_text = !oaicompat_msg.content.empty();
|
||||
size_t num_tool_calls = oaicompat_msg.tool_calls.size();
|
||||
|
||||
// content block indices: thinking (0) -> text (0 or 1) -> tool_use (n+)
|
||||
size_t thinking_block_index = 0;
|
||||
size_t text_block_index = has_thinking ? 1 : 0;
|
||||
|
||||
bool thinking_block_started = false;
|
||||
bool text_block_started = false;
|
||||
bool text_block_started = false;
|
||||
std::unordered_set<size_t> tool_calls_started;
|
||||
|
||||
for (const auto & diff : oaicompat_msg_diffs) {
|
||||
// handle thinking/reasoning content
|
||||
if (!diff.reasoning_content_delta.empty()) {
|
||||
if (!thinking_block_started) {
|
||||
events.push_back({
|
||||
{"event", "content_block_start"},
|
||||
{"data", {
|
||||
{"type", "content_block_start"},
|
||||
{"index", thinking_block_index},
|
||||
{"content_block", {
|
||||
{"type", "thinking"},
|
||||
{"thinking", ""}
|
||||
}}
|
||||
}}
|
||||
});
|
||||
thinking_block_started = true;
|
||||
}
|
||||
|
||||
events.push_back({
|
||||
{"event", "content_block_delta"},
|
||||
{"data", {
|
||||
{"type", "content_block_delta"},
|
||||
{"index", thinking_block_index},
|
||||
{"delta", {
|
||||
{"type", "thinking_delta"},
|
||||
{"thinking", diff.reasoning_content_delta}
|
||||
}}
|
||||
}}
|
||||
});
|
||||
}
|
||||
|
||||
// handle regular text content
|
||||
if (!diff.content_delta.empty()) {
|
||||
if (!text_block_started) {
|
||||
events.push_back({
|
||||
{"event", "content_block_start"},
|
||||
{"data", {
|
||||
{"type", "content_block_start"},
|
||||
{"index", text_block_index},
|
||||
{"index", 0},
|
||||
{"content_block", {
|
||||
{"type", "text"},
|
||||
{"text", ""}
|
||||
@@ -938,7 +886,7 @@ json server_task_result_cmpl_final::to_json_anthropic_stream() {
|
||||
{"event", "content_block_delta"},
|
||||
{"data", {
|
||||
{"type", "content_block_delta"},
|
||||
{"index", text_block_index},
|
||||
{"index", 0},
|
||||
{"delta", {
|
||||
{"type", "text_delta"},
|
||||
{"text", diff.content_delta}
|
||||
@@ -947,9 +895,8 @@ json server_task_result_cmpl_final::to_json_anthropic_stream() {
|
||||
});
|
||||
}
|
||||
|
||||
// handle tool calls
|
||||
if (diff.tool_call_index != std::string::npos) {
|
||||
size_t content_block_index = (has_thinking ? 1 : 0) + (has_text ? 1 : 0) + diff.tool_call_index;
|
||||
size_t content_block_index = (has_text ? 1 : 0) + diff.tool_call_index;
|
||||
|
||||
if (tool_calls_started.find(diff.tool_call_index) == tool_calls_started.end()) {
|
||||
const auto & full_tool_call = oaicompat_msg.tool_calls[diff.tool_call_index];
|
||||
@@ -985,42 +932,18 @@ json server_task_result_cmpl_final::to_json_anthropic_stream() {
|
||||
}
|
||||
}
|
||||
|
||||
// close content blocks in order
|
||||
if (has_thinking) {
|
||||
// Anthropic API requires a signature_delta before closing thinking blocks
|
||||
// We use an empty signature since we can't generate a cryptographic signature for local models
|
||||
events.push_back({
|
||||
{"event", "content_block_delta"},
|
||||
{"data", {
|
||||
{"type", "content_block_delta"},
|
||||
{"index", thinking_block_index},
|
||||
{"delta", {
|
||||
{"type", "signature_delta"},
|
||||
{"signature", ""}
|
||||
}}
|
||||
}}
|
||||
});
|
||||
events.push_back({
|
||||
{"event", "content_block_stop"},
|
||||
{"data", {
|
||||
{"type", "content_block_stop"},
|
||||
{"index", thinking_block_index}
|
||||
}}
|
||||
});
|
||||
}
|
||||
|
||||
if (has_text) {
|
||||
events.push_back({
|
||||
{"event", "content_block_stop"},
|
||||
{"data", {
|
||||
{"type", "content_block_stop"},
|
||||
{"index", text_block_index}
|
||||
{"index", 0}
|
||||
}}
|
||||
});
|
||||
}
|
||||
|
||||
for (size_t i = 0; i < num_tool_calls; i++) {
|
||||
size_t content_block_index = (has_thinking ? 1 : 0) + (has_text ? 1 : 0) + i;
|
||||
size_t content_block_index = (has_text ? 1 : 0) + i;
|
||||
events.push_back({
|
||||
{"event", "content_block_stop"},
|
||||
{"data", {
|
||||
@@ -1228,10 +1151,11 @@ json server_task_result_rerank::to_json() {
|
||||
json server_task_result_cmpl_partial::to_json_anthropic() {
|
||||
json events = json::array();
|
||||
bool first = (n_decoded == 1);
|
||||
// use member variables to track block state across streaming calls
|
||||
// (anthropic_thinking_block_started, anthropic_text_block_started)
|
||||
bool text_block_started = false;
|
||||
|
||||
if (first) {
|
||||
text_block_started = false;
|
||||
|
||||
events.push_back({
|
||||
{"event", "message_start"},
|
||||
{"data", {
|
||||
@@ -1253,69 +1177,28 @@ json server_task_result_cmpl_partial::to_json_anthropic() {
|
||||
});
|
||||
}
|
||||
|
||||
// content block indices: thinking (0) -> text (0 or 1) -> tool_use (n+)
|
||||
size_t thinking_block_index = 0;
|
||||
// use anthropic_has_reasoning (set in update()) to know if ANY reasoning was generated
|
||||
size_t text_block_index = anthropic_has_reasoning ? 1 : 0;
|
||||
|
||||
// use local copies of streaming state (copied from task_result_state in update())
|
||||
// these reflect the state BEFORE this chunk was processed
|
||||
bool thinking_started = anthropic_thinking_block_started;
|
||||
bool text_started = anthropic_text_block_started;
|
||||
|
||||
for (const auto & diff : oaicompat_msg_diffs) {
|
||||
// handle thinking/reasoning content
|
||||
if (!diff.reasoning_content_delta.empty()) {
|
||||
if (!thinking_started) {
|
||||
events.push_back({
|
||||
{"event", "content_block_start"},
|
||||
{"data", {
|
||||
{"type", "content_block_start"},
|
||||
{"index", thinking_block_index},
|
||||
{"content_block", {
|
||||
{"type", "thinking"},
|
||||
{"thinking", ""}
|
||||
}}
|
||||
}}
|
||||
});
|
||||
thinking_started = true;
|
||||
}
|
||||
|
||||
events.push_back({
|
||||
{"event", "content_block_delta"},
|
||||
{"data", {
|
||||
{"type", "content_block_delta"},
|
||||
{"index", thinking_block_index},
|
||||
{"delta", {
|
||||
{"type", "thinking_delta"},
|
||||
{"thinking", diff.reasoning_content_delta}
|
||||
}}
|
||||
}}
|
||||
});
|
||||
}
|
||||
|
||||
// handle regular text content
|
||||
if (!diff.content_delta.empty()) {
|
||||
if (!text_started) {
|
||||
if (!text_block_started) {
|
||||
events.push_back({
|
||||
{"event", "content_block_start"},
|
||||
{"data", {
|
||||
{"type", "content_block_start"},
|
||||
{"index", text_block_index},
|
||||
{"index", 0},
|
||||
{"content_block", {
|
||||
{"type", "text"},
|
||||
{"text", ""}
|
||||
}}
|
||||
}}
|
||||
});
|
||||
text_started = true;
|
||||
text_block_started = true;
|
||||
}
|
||||
|
||||
events.push_back({
|
||||
{"event", "content_block_delta"},
|
||||
{"data", {
|
||||
{"type", "content_block_delta"},
|
||||
{"index", text_block_index},
|
||||
{"index", 0},
|
||||
{"delta", {
|
||||
{"type", "text_delta"},
|
||||
{"text", diff.content_delta}
|
||||
@@ -1324,10 +1207,8 @@ json server_task_result_cmpl_partial::to_json_anthropic() {
|
||||
});
|
||||
}
|
||||
|
||||
// handle tool calls
|
||||
if (diff.tool_call_index != std::string::npos) {
|
||||
// use anthropic_has_reasoning for thinking block count (persists across calls)
|
||||
size_t content_block_index = (anthropic_has_reasoning ? 1 : 0) + (text_started ? 1 : 0) + diff.tool_call_index;
|
||||
size_t content_block_index = (text_block_started ? 1 : 0) + diff.tool_call_index;
|
||||
|
||||
if (!diff.tool_call_delta.name.empty()) {
|
||||
events.push_back({
|
||||
|
||||
Reference in New Issue
Block a user