Skip to content
llm-speed
Leaderboard/RTX 4090/Qwen2.5-Coder-14B-Instruct

Can an RTX 4090 run Qwen2.5-Coder-14B-Instruct?

llm-speed has not measured Qwen2.5-Coder-14B-Instruct on RTX 4090 itself yet, so there is no llm-speed verdict. 2 community-submitted runs exist (fastest 89.9 tok/s at Q4_K_M on llama.cpp); community-submitted, not measured by llm-speed, so they give no verdict.

2 signed runs on llm-speed. Single-request decode speed in tokens per second; the short-chat workload unless noted. All 2 runs are community-submitted (signed on the submitter's own machine, not measured by llm-speed, and not used for llm-speed verdicts). Verdicts and headline numbers use only llm-speed's own runs and fit tests.

Measured tok/s by quant and runtime

QuantRuntimeBuildDecode tok/s (best run)WorkloadRunsSource
Q4_K_Mllama.cppcommit df18bee89.9chat-short1Community-submitted
Q4_K_Mollama0.31.181.2chat-short1Community-submitted

llm-speed has not measured this pair itself, so there is no llm-speed verdict or fastest quant. The rows above are community-submitted, not measured by llm-speed.

VRAM: not recorded for these runs. The VRAM calculator gives an estimate (predicted, not measured).

Every measured row

RunDate (UTC)WorkloadQuantRuntime / buildPrompt tokensOutput tokensDecode tok/sPrefill tok/sTTFT msVRAM GiBSource
r_eqnsqmgz29a2026-10-14chat-shortQ4_K_Mllama.cpp · commit df18bee13125689.92292.657–Community-submitted
r_ttniv2j74vk2026-10-14chat-shortQ4_K_Mollama · 0.31.113125681.2324.4404–Community-submitted
r_eqnsqmgz29a2026-10-14chat-longQ4_K_Mllama.cpp · commit df18bee316842684.24989.1635–Community-submitted
r_ttniv2j74vk2026-10-14chat-longQ4_K_Mollama · 0.31.1316860376.63252.2974–Community-submitted
r_eqnsqmgz29a2026-10-14agent-traceQ4_K_Mllama.cpp · commit df18bee326884086.614684.0142–Community-submitted
r_ttniv2j74vk2026-10-14agent-traceQ4_K_Mollama · 0.31.1326884078.64874.6449–Community-submitted
r_eqnsqmgz29a2026-10-14concurrent-decodeQ4_K_Mllama.cpp · commit df18bee102425688.9–––Community-submitted
r_ttniv2j74vk2026-10-14concurrent-decodeQ4_K_Mollama · 0.31.1102425680.1–––Community-submitted

Questions

Can an RTX 4090 run Qwen2.5-Coder-14B-Instruct?

llm-speed has not measured Qwen2.5-Coder-14B-Instruct on RTX 4090 itself yet, so there is no llm-speed verdict. 2 community-submitted runs exist (fastest 89.9 tok/s at Q4_K_M on llama.cpp); community-submitted, not measured by llm-speed, so they give no verdict.

What did community runs measure for Qwen2.5-Coder-14B-Instruct on an RTX 4090?

2 community-submitted runs (community-submitted, not measured by llm-speed): Q4_K_M on llama.cpp: 89.9 tok/s; Q4_K_M on ollama: 81.2 tok/s. They are listed for reference and give no llm-speed verdict.

Sources and related pages