met 1.0.0-beta-preview

Metrale Engine: pure Rust LLM inference server (HTTP API, scheduler, TUI, CLI).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133

// SPDX-License-Identifier: MIT OR Apache-2.0

//! 2026-09-26: Streaming branch of `chat_completions_inner` (`chat/mod.rs`):
//! refuses `n > 1`, then hands the request to `run_chat_stream`.
//!
//! Owner: server chat API.
//! Invariants: none beyond the types.

#![allow(clippy::too_many_arguments)]

use std::sync::Arc;

use axum::http::StatusCode;

use crate::AppState;
use crate::ir::ChatRequest;
use crate::tool_parser;

use super::chat_stream::run_chat_stream;
use super::compact::openai_error_response;
use super::inference_types::GrammarSpec;

pub(super) async fn dispatch_streaming(
    state: Arc<AppState>,
    req: &ChatRequest,
    req_ctx: Option<axum::extract::Extension<crate::rate_limiter::RequestContext>>,
    dump_seq: Option<u64>,
    prompt_tokens: Vec<u32>,
    session_hash: u64,
    // 2026-09-26: Resolved LoRA adapter slot; -1 follows the active adapter.
    adapter_slot: i32,
    // 2026-09-26: Source-language token id; 0 means the deployment default.
    src_lang_id: u32,
    // 2026-09-26: Target-language token id; 0 means the deployment default.
    tgt_lang_id: u32,
    // 2026-09-26: NLLB beam-search parameters. `chat_completions_inner` refuses
    // `num_beams > 1` with streaming before this call.
    num_beams: u32,
    length_penalty: f32,
    early_stopping: bool,
    image_pixels: Vec<metrale_model_layers::VisionItem>,
    max_tokens: usize,
    temperature: f32,
    top_k: u32,
    top_p: f32,
    top_n_sigma: f32,
    min_p: f32,
    repetition_penalty: f32,
    presence_penalty: f32,
    frequency_penalty: f32,
    dry_multiplier: f32,
    dry_base: f32,
    dry_allowed_length: u32,
    lz_penalty: f32,
    logit_bias: Vec<(u32, f32)>,
    enable_thinking: bool,
    thinking_budget: Option<u32>,
    tools_active: bool,
    tool_choice_required: bool,
    suppress_tool_call: bool,
    cwd_hint: Option<String>,
    stop_tokens: Vec<u32>,
    grammar_spec: Option<GrammarSpec>,
    top_logprobs: Option<u8>,
    timeout_at: Option<std::time::Instant>,
    active_guard: crate::metrics::ActiveRequestGuard,
) -> super::chat::ChatOutcome {
    if req.n > 1 {
        // 2026-09-26: Returning drops `active_guard`, which decrements
        // `REQUESTS_ACTIVE`.
        return super::chat::ChatOutcome::Http(openai_error_response(
            StatusCode::BAD_REQUEST,
            "n > 1 is not supported in streaming mode".to_string(),
        ));
    }
    let tool_defs: Vec<tool_parser::ToolDefinition> = req.tools.clone();
    let mut stop_strings = req.stop.clone();
    stop_strings.sort_by_key(|s| std::cmp::Reverse(s.len()));
    let req_return_token_ids = req.return_token_ids;
    let ctx_for_stream = req_ctx.as_ref().map(|e| e.0.clone());
    let repetition_detection = req.repetition_detection;
    match run_chat_stream(
        state,
        prompt_tokens,
        session_hash,
        req.lookup_tenant,
        adapter_slot,
        src_lang_id,
        tgt_lang_id,
        num_beams,
        length_penalty,
        early_stopping,
        image_pixels,
        max_tokens,
        req.min_tokens,
        temperature,
        top_k,
        top_p,
        top_n_sigma,
        min_p,
        repetition_penalty,
        presence_penalty,
        frequency_penalty,
        dry_multiplier,
        dry_base,
        dry_allowed_length,
        lz_penalty,
        logit_bias,
        enable_thinking,
        thinking_budget,
        repetition_detection,
        tools_active,
        tool_choice_required,
        suppress_tool_call,
        tool_defs,
        cwd_hint,
        stop_tokens,
        grammar_spec,
        req.seed,
        top_logprobs,
        timeout_at,
        stop_strings,
        req_return_token_ids,
        ctx_for_stream,
        dump_seq,
        active_guard,
    )
    .await
    {
        Ok(deltas) => super::chat::ChatOutcome::Streaming(deltas),
        Err((status, msg)) => super::chat::ChatOutcome::Http(openai_error_response(status, msg)),
    }
}