LiteLLM · Python vs Bifrost

Operated by the Busbar team; Busbar is also an entrant. Gateway overhead against a local mock is not end-to-end model latency. Methodology and limitations.

Requested cell: OpenAI → OpenAI. Metric: Added latency p99 (µs).

Comparability: Comparable. Ratios require an identical protocol pair, hardware, architecture, metric definition, clean harness commit, recorded mock digest, and metric-specific workload. Missing evidence never matches missing evidence. These are observed gateway measurements against a local mock, not a claim about model latency or statistical significance.

Ratio (Bifrost ÷ LiteLLM · Python): 0.111. No statistical significance is asserted.

Comparison JSON · All LiteLLM · Python evidence · All Bifrost evidence

LiteLLM · Python

Build ghcr.io/berriai/litellm:v1.108.3. Measured . Run evidence

Ingress → upstreamMetricValueState / reasonp99 bound (ms)
OpenAI → OpenAIAdded latency p50 (µs)8,512Valid · n=701 gateway / 267,297 direct—
OpenAI → OpenAIAdded latency p99 (µs)9,522Valid · n=701 gateway / 267,297 direct—
OpenAI → OpenAIGateway p99 at concurrency 1 (µs)9,549Valid · n=701—
OpenAI → OpenAIDirect-to-mock p99 at concurrency 1 (µs)27Valid · n=267,297—
OpenAI → OpenAICPU per request (µs)8,644.8Valid—
OpenAI → OpenAIRequests per CPU-second115.7Valid—
OpenAI → OpenAICost window concurrency8Valid—
OpenAI → OpenAICost window core utilisation0.25Valid—
OpenAI → OpenAICost window successful requests701Valid—
OpenAI → OpenAICost window throughput (req/s)116Valid—
OpenAI → OpenAICost window threads32Valid—
OpenAI → OpenAINon-voluntary context switches per request0.029Valid—
OpenAI → OpenAICost window major page faults0Valid—
OpenAI → OpenAIAdded time to first event p50 (µs)11,154Valid · n=100 gateway / 100 direct—
OpenAI → OpenAIAdded time to first event p99 (µs)11,974Valid · n=100 gateway / 100 direct—
OpenAI → OpenAIAdded inter-frame gap p50 (µs)—Below resolution · n=63 gateway / 63 direct—
OpenAI → OpenAIAdded inter-frame gap p99 (µs)9Valid · n=63 gateway / 63 direct—
OpenAI → OpenAIStreams sustained (frames/s)1,208Valid—
OpenAI → OpenAIStreams sustained62Valid—
OpenAI → OpenAIIdle RSS (MiB)593Valid—
OpenAI → OpenAISteady-state RSS (MiB)628.6Valid—
OpenAI → OpenAIRecovered RSS (MiB)628.6Valid—
OpenAI → OpenAIPeak RSS (MiB)628.6Valid—
OpenAI → OpenAIPeak RSS high-water mark (MiB)628.6Valid—
OpenAI → OpenAITime to plateau (s)61Valid—
OpenAI → OpenAIMemory growth (MiB/min)0Valid—
OpenAI → OpenAIMemory load duration (s)302Valid—
OpenAI → OpenAIMemory shape under load—Not measured—
OpenAI → OpenAIIdle memory shape—Not measured—
OpenAI → OpenAIIdle memory static1Valid—
OpenAI → OpenAIIdle memory growth (MiB/min)0Valid—
OpenAI → OpenAIThroughput (req/s)—Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound1 ms
OpenAI → OpenAIThroughput (req/s)—Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 5ms, so this gateway carried no measurable throughput under that bound5 ms
OpenAI → OpenAIThroughput (req/s)117concurrency 1; n=70810 ms
OpenAI → OpenAIThroughput (req/s)119concurrency 2; n=71850 ms
OpenAI → OpenAIThroughput (req/s)119concurrency 2; n=718100 ms
OpenAI → OpenAIThroughput (req/s)119concurrency 2; n=718Unbounded

Bifrost

Build maximhq/bifrost:v2.5.0. Measured . Run evidence

Ingress → upstreamMetricValueState / reasonp99 bound (ms)
OpenAI → OpenAIAdded latency p50 (µs)658Valid · n=8,543 gateway / 269,310 direct—
OpenAI → OpenAIAdded latency p99 (µs)1,057Valid · n=8,543 gateway / 269,310 direct—
OpenAI → OpenAIGateway p99 at concurrency 1 (µs)1,084Valid · n=8,543—
OpenAI → OpenAIDirect-to-mock p99 at concurrency 1 (µs)27Valid · n=269,310—
OpenAI → OpenAICPU per request (µs)865.1Valid—
OpenAI → OpenAIRequests per CPU-second1,155.9Valid—
OpenAI → OpenAICost window concurrency8Valid—
OpenAI → OpenAICost window core utilisation0.991Valid—
OpenAI → OpenAICost window successful requests27,580Valid—
OpenAI → OpenAICost window throughput (req/s)4,595Valid—
OpenAI → OpenAICost window threads13Valid—
OpenAI → OpenAINon-voluntary context switches per request0.027Valid—
OpenAI → OpenAICost window major page faults0Valid—
OpenAI → OpenAIAdded time to first event p50 (µs)932Valid · n=100 gateway / 100 direct—
OpenAI → OpenAIAdded time to first event p99 (µs)1,417Valid · n=100 gateway / 100 direct—
OpenAI → OpenAIAdded inter-frame gap p50 (µs)3Valid · n=63 gateway / 63 direct—
OpenAI → OpenAIAdded inter-frame gap p99 (µs)—Below resolution · n=63 gateway / 63 direct—
OpenAI → OpenAIStreams sustained (frames/s)25,840.9Valid—
OpenAI → OpenAIStreams sustained580Valid—
OpenAI → OpenAIIdle RSS (MiB)160.3Valid—
OpenAI → OpenAISteady-state RSS (MiB)1,049.5Valid—
OpenAI → OpenAIRecovered RSS (MiB)1,256.6Valid—
OpenAI → OpenAIPeak RSS (MiB)1,257Valid—
OpenAI → OpenAIPeak RSS high-water mark (MiB)1,137.4Valid—
OpenAI → OpenAITime to plateau (s)—Not measured—
OpenAI → OpenAIMemory growth (MiB/min)56.82Valid—
OpenAI → OpenAIMemory load duration (s)300Valid—
OpenAI → OpenAIMemory shape under load1Valid—
OpenAI → OpenAIIdle memory shape—Not measured—
OpenAI → OpenAIIdle memory static1Valid—
OpenAI → OpenAIIdle memory growth (MiB/min)0.48Valid—
OpenAI → OpenAIThroughput (req/s)—Bound not met; concurrency not recorded: every cleanly-served rung that reported a tail had it at or above 1ms, so this gateway carried no measurable throughput under that bound1 ms
OpenAI → OpenAIThroughput (req/s)4,346concurrency 4; n=26,0815 ms
OpenAI → OpenAIThroughput (req/s)4,555concurrency 8; n=27,34010 ms
OpenAI → OpenAIThroughput (req/s)4,624concurrency 16; n=27,74850 ms
OpenAI → OpenAIThroughput (req/s)4,624concurrency 16; n=27,748100 ms
OpenAI → OpenAIThroughput (req/s)4,624concurrency 16; n=27,748Unbounded