Qwen-DeepSeek 1.5B Agentic Distill

model page · concurrency sweep 1–24 agents

Qwen-DeepSeek 1.5B Agentic Distill

server model id: qwen-deepseek-1.5b-agentic-distill
thinkingKV F16ctx 32k1.5B params

Best scaling efficiency (5.0x) but the worst TTFT profile — 20–58s to first token under load.

Key findings

  • Strong late-game scaling: combined throughput climbs steadily to 829 tok/s at 21 agents (5.0x) and holds 726–774 at 22–24 — the highest scaling of any model tested so far. A jump at 21+ likely reflects LM Studio batching kicking in more aggressively.
  • Visible content is modest: content-only throughput peaks at only ~299 tok/s (16 agents) — over half the budget is hidden reasoning_content (31k–34k reasoning tokens per run at high concurrency).
  • Very slow TTFT — the worst profile: 4.2s solo, and 20–36s mean under load, with max values up to 58s (near the full 60s window). Agents frequently wait half the run before their first token. Bad for latency-critical use.
  • Per-agent speed: ~30–50 tok/s (all-token) at high concurrency.
  • Errors: 0 across all 24 runs.

Sweep — total concurrency 1–24

agentswall (s)contentreasontotalcomb tok/scomb allper-agentper-agent allscale %TTFT meanTTFT maxoktimeouterr
160.06,3873,5289,915106165654.1459.7100%4.2s4.2s010
260.012,5663,54416,110209268929.8469.4162%10.6s18.2s020
360.011,0595,75316,812184280106.4153.3170%9.2s11.0s030
460.05,88916,08821,9779836670.8214.6222%10.6s15.5s040
560.011,68010,49522,175195369124.8145.4224%20.7s47.6s050
660.111,44410,26021,70419036159.670.9219%14.1s27.6s060
760.011,96510,74022,70519937858.083.7229%18.5s28.5s070
860.014,2718,96523,23623838759.070.8235%20.4s45.2s080
960.016,30810,22526,53327244263.777.7268%17.0s23.3s090
1060.012,47715,46427,94120846553.385.9282%21.5s31.5s0100
1160.014,41714,90729,32424048847.167.7296%24.6s37.1s0110
1260.012,71615,38728,10321246852.677.4284%26.4s57.9s0120
1360.015,02715,88830,91525051545.666.1312%30.4s48.5s0130
1460.014,59014,37728,96724348241.654.2292%27.4s49.6s0140
1560.010,28820,86031,14817151932.255.9315%32.1s49.9s0150
1660.017,95814,48432,44229954039.649.8327%31.2s49.8s0160
1760.012,84718,51231,35921452234.249.2316%35.9s48.0s0170
1860.014,86117,33832,19924853632.746.6325%30.2s37.0s0180
1960.015,57017,25432,82425954737.054.4332%35.2s52.9s0190
2060.013,69718,18431,88122853129.745.6322%33.1s47.8s0200
21 peak60.016,05333,73149,78426782938.065.1502%23.4s50.2s0210
2260.017,57626,03343,60929372671.659.2440%26.8s41.0s0220
2360.017,31129,15746,46828877450.263.8469%27.9s51.6s0230
2460.014,58831,24345,83124376339.663.7462%27.8s54.5s0240

* 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