Chronos 1.5B

model page · concurrency sweep 1–24 agents

Chronos 1.5B

server model id: chronos-1.5b
thinkingKV F16ctx 8k1.5B params

Slowest solo (88 tok/s) but scales 5.2x — 94% of its output at high concurrency is hidden reasoning.

Key findings

  • The slowest model tested: only 88 tok/s solo (content 78), and combined all-token throughput tops out at just 458 tok/s at 24 agents — far below every other model (llama 652, falcon3 839, qwen-deepseek 829, minicpm 983).
  • Impressive relative scaling (520% at 23–24) — but that's mostly because the single-agent baseline is so low.
  • Reasoning dominates under load: content-only throughput is tiny at high concurrency (26 tok/s at 24) — ~94% of output is hidden reasoning_content (25k–26k reasoning tokens per run at 22–24 agents).
  • TTFT degrades steadily: 226ms solo → 17–18s at 20–24 agents.
  • Errors: 0 across all 24 runs.

Sweep — total concurrency 1–24

agentswall (s)contentreasontotalcomb tok/scomb allper-agentper-agent allscale %TTFT meanTTFT maxoktimeouterr
160.04,7035655,268788878.788.1100%226ms226ms010
260.01,0428,4429,484171588.779.2180%130ms144ms020
360.01,71610,28311,9992920010.472.5227%4.8s4.9s030
460.07,8435,63913,48213122538.766.9256%7.0s7.0s040
560.06,59310,09116,68411027825.363.7316%4.5s4.5s050
660.04,54613,74218,2887630514.759.0347%8.3s8.4s060
760.05,18114,39119,5728632614.956.4370%10.4s10.4s070
860.01,58918,58320,172263364.752.8382%11.0s11.1s080
960.05,45316,61422,0679136812.148.8418%9.8s9.8s090
1060.04,15018,44322,593693768.647.1427%12.0s12.0s0100
1160.03,26220,59323,855543976.245.1451%11.9s11.9s0110
1260.03,54520,36823,913593986.443.2452%13.8s13.9s0120
1360.06,79617,12123,91711339811.941.3452%15.1s15.1s0130
1460.05,64219,62925,271944218.940.0478%14.8s14.9s0140
1560.04,40220,90225,304734216.638.2478%15.8s15.8s0150
1660.04,57721,15025,727764296.536.8488%16.3s16.4s0160
1760.02,88021,88224,762484123.833.0468%15.8s15.9s0170
1860.02,19423,66725,861374312.732.4490%15.7s15.7s0180
1960.02,94723,27726,224494373.531.7497%16.5s16.6s0190
2060.03,08122,89125,972514333.630.7492%17.7s17.8s0200
2160.01,39124,76226,153234361.629.7495%18.1s18.2s0210
2260.02,50824,74327,251424542.628.9516%17.1s17.2s0220
23 peak60.02,97124,55527,526494583.028.1520%17.3s17.5s0230
2460.01,57725,91727,494264581.526.8520%17.3s17.4s0240

* burst artifact — agents queued ~11–16s then generated in a short burst; the per-agent timer inflates the number. Trust combined throughput and TTFT columns. Peak row = highest combined (all-token) throughput.

Charts

Whole sweep at a glance — combined throughput, scaling efficiency, per-agent throughput and outcomes across every concurrency level.
Whole sweep at a glance — combined throughput, scaling efficiency, per-agent throughput and outcomes across every concurrency level.
Combined throughput (all tokens) vs concurrent agents. Annotated peak with the concurrency level where it occurred.
Combined throughput (all tokens) vs concurrent agents. Annotated peak with the concurrency level where it occurred.
Per-agent throughput erosion as slots contend. Red stars mark burst-timing artifacts (queued ≥ 10s then generated in a burst — inflated values, not real throughput).
Per-agent throughput erosion as slots contend. Red stars mark burst-timing artifacts (queued ≥ 10s then generated in a burst — inflated values, not real throughput).
Scaling efficiency relative to the single-agent baseline (100%).
Scaling efficiency relative to the single-agent baseline (100%).
Time to first token — mean (solid) and max (dashed). Log scale. Gaps mean no visible content ever arrived within the 60s window.
Time to first token — mean (solid) and max (dashed). Log scale. Gaps mean no visible content ever arrived within the 60s window.
Total tokens per 60s run — all tokens vs visible content only.
Total tokens per 60s run — all tokens vs visible content only.
How each run of N agents finished: completed, timed out (60s budget), or errored.
How each run of N agents finished: completed, timed out (60s budget), or errored.
The concurrency trade-off on one chart — combined throughput rises while per-agent speed falls.
The concurrency trade-off on one chart — combined throughput rises while per-agent speed falls.
Visible content vs hidden reasoning tokens per run, plus the reasoning share (%).
Visible content vs hidden reasoning tokens per run, plus the reasoning share (%).

Raw data & downloads