Fix Responses API background=true + stream=true panic (#1945)

[glaziermag]

Fixes #1945. Alternative to closed PR #1989.

Problem

The /v1/responses endpoint supports a background=true mode that spawns a Tokio task to process the request asynchronously. When stream=true was also set, the pipeline emitted a sequence of Response::Chunk(…) messages followed by a terminal Response::Done. The background receiver used a single match bg_rx.recv().await — it hit the first Chunk message, fell into the _ => catch-all, and marked the task as failed with "Unexpected response type".

Fix

Replace the single match with a while let Some(msg) = bg_rx.recv().await loop. Response::Chunk(_) arms continue to drain stream chunks; the terminal variants (Done, ModelError, ValidationError, InternalError) break the loop with the appropriate outcome.

Scope

This fix applies to background (background=true) chat completion requests. Non-background streaming requests go through a different path and are unaffected. Other response variants (CompletionDone, ImageGeneration, etc.) not applicable to this endpoint fall into the _ => arm as before — that is pre-existing behavior for endpoint misuse, not a regression.

Files changed

• mistralrs-server-core/src/responses.rs

[github-actions]

Code Metrics Report

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 Language              Files        Lines         Code     Comments       Blanks
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 C Header                  5          305          210           52           43
 CSS                       2         1181         1036           34          111
 CUDA                     59        17706        13869         1637         2200
 Dockerfile                1           39           22            8            9
 HTML                      2          235          197           14           24
 JavaScript               16         3580         2702          486          392
 Jinja2                    7          694          656            5           33
 JSON                     21          409          406            0            3
 Makefile                  1            6            5            0            1
 Metal Shading Lan|       31        11647         9007         1064         1576
 PowerShell                1          300          227           30           43
 Python                  125         8316         6808          412         1096
 Shell                     2          485          329           95           61
 Plain Text                3         3723            0         2413         1310
 TOML                     27         1290         1124           35          131
 YAML                      3           25           23            2            0
─────────────────────────────────────────────────────────────────────────────────
 Jupyter Notebooks         3          122           83           23           16
 |- Markdown               1           60           30           22            8
 |- Python                 1          122          113            1            8
 (Total)                              304          226           46           32
─────────────────────────────────────────────────────────────────────────────────
 Markdown                105        11197            0         8067         3130
 |- BASH                  72          934          691          149           94
 |- Dockerfile             1            1            1            0            0
 |- JSON                  20          719          719            0            0
 |- PowerShell             3            3            3            0            0
 |- Python                23         1038          862           60          116
 |- Rust                  51         2048         1718           54          276
 |- TOML                   6          207          164            0           43
 |- YAML                   2            9            8            1            0
 (Total)                            16156         4166         8331         3659
─────────────────────────────────────────────────────────────────────────────────
 Rust                    547       236072       207590         6565        21917
 |- Markdown             361         8962          452         7385         1125
 (Total)                           245034       208042        13950        23042
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 Total                   961       311435       249055        28614        33766
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

[glaziermag]

Update (2026-04-15): Rebased onto origin/master. Branch was previously behind upstream by several commits. No code changes — the while let Some(msg) = bg_rx.recv().await fix with Response::Chunk(_) => continue is unchanged.

[glaziermag]

Closing in favor of #2114, which is a clean reimplementation against current master with a minimal diff (no extraneous reindentation changes). The fix is the same: loop-drain Response::Chunk events in the background task handler.