[ImgBot] Optimize images #1
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ting at Scale Signed-off-by: bitliu <[email protected]>
*Total -- 62,985.50kb -> 51,682.67kb (17.95%) /assets/figures/semantic-router/signal-5.png -- 392.85kb -> 98.53kb (74.92%) /assets/figures/semantic-router/signal-2.png -- 438.48kb -> 111.89kb (74.48%) /assets/figures/semantic-router/signal-7.png -- 190.01kb -> 51.13kb (73.09%) /assets/figures/semantic-router/signal-1.png -- 136.21kb -> 38.90kb (71.44%) /assets/figures/semantic-router/signal-6.png -- 172.88kb -> 50.59kb (70.74%) /assets/figures/semantic-router/signal-3.png -- 129.68kb -> 40.34kb (68.89%) /assets/figures/annimation2.gif -- 183.06kb -> 63.14kb (65.51%) /assets/figures/minimax-m1/lightning_attention.png -- 137.93kb -> 52.24kb (62.12%) /assets/figures/vllm-2024-wrapped-2025-roadmap/gpu-hours-by-vendor.png -- 155.92kb -> 59.38kb (61.92%) /assets/figures/2025-torch-compile/figure3.png -- 158.74kb -> 61.10kb (61.51%) /assets/figures/2025-11-10-bitwise-exact-rl/floating-point-representation.png -- 58.85kb -> 23.65kb (59.82%) /assets/figures/2025-vllm-anatomy/server_setup.png -- 41.53kb -> 16.85kb (59.42%) /assets/figures/2025-torch-compile/figure4.png -- 120.94kb -> 50.23kb (58.46%) /assets/figures/2025-torch-compile/figure2.png -- 104.33kb -> 43.59kb (58.22%) /assets/figures/vllm-2024-wrapped-2025-roadmap/quantization-deployment-percentage.png -- 238.96kb -> 103.63kb (56.63%) /assets/figures/2025-torch-compile/figure7.png -- 35.34kb -> 15.36kb (56.53%) /assets/figures/2025-torch-compile/figure6.png -- 177.58kb -> 77.74kb (56.22%) /assets/figures/semantic-router/signal-4.png -- 191.29kb -> 87.34kb (54.34%) /assets/logos/vllm-logo-only-light.png -- 52.94kb -> 24.32kb (54.05%) /assets/figures/2025-torch-compile/figure8.png -- 16.32kb -> 7.77kb (52.4%) /assets/figures/semantic-router/signal-8.png -- 263.82kb -> 127.87kb (51.53%) /assets/figures/blackwell-inferencemax/gpt-oss-120b-1k-1k.png -- 366.50kb -> 180.69kb (50.7%) /assets/figures/blackwell-inferencemax/llama-70b-1k-8k.png -- 366.67kb -> 181.65kb (50.46%) /assets/figures/spec-decode/figure2.png -- 198.98kb -> 101.82kb (48.83%) /assets/figures/2025-shm-ipc-cache/processes2.png -- 160.40kb -> 82.54kb (48.55%) /assets/figures/struct-decode-intro/shogoth-gpt.png -- 53.86kb -> 27.74kb (48.51%) /assets/figures/2025-shm-ipc-cache/processes1.png -- 144.34kb -> 74.98kb (48.05%) /assets/figures/llama4/perf.png -- 514.08kb -> 268.36kb (47.8%) /assets/logos/vllm-logo-text-light.png -- 87.87kb -> 45.92kb (47.74%) /assets/figures/vllm-tpu/llama3-8b-throughput-progress.png -- 47.28kb -> 24.75kb (47.65%) /assets/figures/perf-v060/illustration-async-output-processing.png -- 189.81kb -> 99.47kb (47.6%) /assets/figures/vllm-tpu/llama3-70b-throughput-progress.png -- 43.75kb -> 23.12kb (47.15%) /assets/figures/spec-decode/figure3.png -- 147.58kb -> 78.34kb (46.92%) /assets/figures/annimation3.gif -- 449.13kb -> 240.12kb (46.54%) /assets/figures/vllm-2024-wrapped-2025-roadmap/model-architecture-serving-usage.png -- 209.66kb -> 112.70kb (46.25%) /assets/logos/vllm-logo-text-dark.png -- 86.27kb -> 46.79kb (45.76%) /assets/figures/2025-shm-ipc-cache/shared_memory_object_store.png -- 247.46kb -> 134.76kb (45.54%) /assets/figures/perf-v060/illustration-multi-step.png -- 216.43kb -> 118.06kb (45.45%) /assets/figures/deepseek-v3-2/dsa-explained.png -- 1,178.67kb -> 663.26kb (43.73%) /assets/figures/distributed-inference/kv_cache_effects.png -- 85.85kb -> 48.72kb (43.25%) /assets/figures/perf-v060/illustration-api-server.png -- 194.16kb -> 112.27kb (42.18%) /assets/figures/v1/v1_tp_architecture.png -- 172.54kb -> 101.53kb (41.16%) /assets/figures/v1/v1_scheduling.png -- 133.57kb -> 79.02kb (40.84%) /assets/figures/spec-decode/figure8.png -- 168.19kb -> 100.25kb (40.39%) /assets/figures/2025-vllm-anatomy/fsm.png -- 137.98kb -> 84.14kb (39.02%) /assets/figures/beyond-text/models-diff.png -- 381.34kb -> 239.89kb (37.09%) /assets/figures/annimation1.gif -- 342.45kb -> 216.73kb (36.71%) /assets/figures/v1/vLLM_V1_Logo.png -- 37.17kb -> 23.76kb (36.09%) /assets/figures/lmsys_traffic.png -- 196.11kb -> 125.48kb (36.01%) /assets/figures/2025-11-10-bitwise-exact-rl/tensorboard-plot.png -- 634.28kb -> 410.35kb (35.3%) /assets/figures/spec-decode/figure10.png -- 77.45kb -> 50.38kb (34.96%) /assets/figures/2025-11-10-bitwise-exact-rl/rl-script-demo.png -- 667.21kb -> 437.30kb (34.46%) /assets/figures/2025-11-10-bitwise-exact-rl/reward-comparison.png -- 577.02kb -> 378.58kb (34.39%) /assets/figures/struct-decode-intro/vllm-xgrammar-decode-time-per-output-token.png -- 66.28kb -> 43.66kb (34.12%) /assets/figures/vllm-tpu/vllm-tpu.png -- 189.11kb -> 125.13kb (33.83%) /assets/figures/vllm-serving-amd/case05-amd-recommended-environmental-variables/Mean TPOT (ms).png -- 31.83kb -> 21.10kb (33.71%) /assets/figures/v1/persistent_batch.png -- 78.32kb -> 52.14kb (33.43%) /assets/figures/2025-vllm-anatomy/prefix_pt2.png -- 182.58kb -> 122.45kb (32.93%) /assets/figures/2025-vllm-anatomy/engine_constructor.png -- 185.46kb -> 124.41kb (32.92%) /assets/figures/2025-vllm-anatomy/chunked_pt1.png -- 205.58kb -> 138.92kb (32.43%) /assets/figures/vllm-serving-amd/case05-amd-recommended-environmental-variables/Requests Per Second.png -- 35.06kb -> 23.71kb (32.36%) /assets/figures/2025-vllm-anatomy/pd.png -- 466.79kb -> 317.95kb (31.89%) /assets/figures/2025-vllm-anatomy/prefix_pt3.png -- 335.28kb -> 229.30kb (31.61%) /assets/figures/2025-vllm-anatomy/multiprocexecutor.png -- 77.19kb -> 52.85kb (31.52%) /assets/figures/minimax-m1/moe.png -- 153.55kb -> 105.18kb (31.5%) /assets/figures/agent-lightning/3_agl.png -- 490.02kb -> 337.65kb (31.09%) /assets/figures/2025-vllm-anatomy/latency_diagram.png -- 36.55kb -> 25.20kb (31.05%) /assets/figures/ptpc/PTPC-Diagram.png -- 58.00kb -> 40.03kb (30.98%) /assets/figures/stack/stack-table.png -- 76.69kb -> 52.99kb (30.9%) /assets/figures/vllm-serving-amd/case02-num-scheduler-steps/Requests per Second.png -- 30.30kb -> 20.98kb (30.76%) /assets/figures/v1/torch_compile_cuda_graph.png -- 116.47kb -> 80.89kb (30.55%) /assets/figures/vllm-serving-amd/case03-chunked-prefill-and-prefix-caching/Requests per Second.png -- 46.36kb -> 32.27kb (30.4%) /assets/figures/2025-vllm-nvidia-nemotron/figure1.png -- 275.57kb -> 191.96kb (30.34%) /assets/figures/vllm-serving-amd/case03-chunked-prefill-and-prefix-caching/Mean TPOT (ms).png -- 38.33kb -> 26.80kb (30.08%) /assets/figures/stack/stack-ttft.png -- 80.26kb -> 56.55kb (29.54%) /assets/figures/vllm-serving-amd/case02-num-scheduler-steps/Mean TPOT (ms).png -- 25.92kb -> 18.35kb (29.19%) /assets/figures/spec-decode/figure9.png -- 270.40kb -> 192.07kb (28.97%) /assets/figures/2025-vllm-anatomy/fsm2.png -- 125.90kb -> 90.11kb (28.43%) /assets/figures/stack/stack-itl.png -- 73.40kb -> 52.56kb (28.39%) /assets/figures/2025-vllm-anatomy/kv_cache_blocks.png -- 132.97kb -> 95.24kb (28.37%) /assets/figures/qwen3-next/qwen.png -- 126.26kb -> 90.69kb (28.17%) /assets/figures/2025-vllm-anatomy/specdec_pt1.png -- 177.17kb -> 127.33kb (28.13%) /assets/figures/agent-lightning/1_rewards.png -- 114.37kb -> 82.21kb (28.11%) /assets/figures/vllm-serving-amd/case04-max-seq-len-to-capture/Requests per Second.png -- 53.67kb -> 38.97kb (27.38%) /assets/figures/beyond-text/io-plugins-flow.png -- 205.52kb -> 149.43kb (27.29%) /assets/figures/vllm-serving-amd/case05-amd-recommended-environmental-variables/Mean TTFT (ms).png -- 28.33kb -> 20.62kb (27.23%) /assets/figures/2025-vllm-anatomy/dpenginecoreproc.png -- 189.71kb -> 138.48kb (27%) /assets/figures/spec-decode/figure6.png -- 206.27kb -> 150.69kb (26.95%) /assets/figures/2025-vllm-anatomy/engine_loop.png -- 83.08kb -> 60.91kb (26.69%) /assets/figures/notes-vllm-vs-deepspeed/s2.png -- 33.97kb -> 24.92kb (26.64%) /assets/figures/2025-11-10-bitwise-exact-rl/bf16-rounding-example.png -- 214.02kb -> 157.31kb (26.5%) /assets/figures/vllm-serving-amd/case03-chunked-prefill-and-prefix-caching/Mean TTFT (ms).png -- 35.73kb -> 26.37kb (26.2%) /assets/figures/spec-decode/figure5.png -- 226.69kb -> 167.42kb (26.15%) /assets/figures/notes-vllm-vs-deepspeed/s1.png -- 38.58kb -> 28.54kb (26.01%) /assets/figures/vllm-serving-amd/case04-max-seq-len-to-capture/Mean TPOT (ms).png -- 46.08kb -> 34.10kb (26%) /assets/figures/vllm-serving-amd/case01-chunked-prefill/Mean TPOT (ms).png -- 22.04kb -> 16.34kb (25.88%) /assets/figures/v1/v1_server_architecture.png -- 299.93kb -> 222.39kb (25.85%) /assets/figures/2025-vllm-anatomy/roofline.png -- 47.72kb -> 35.52kb (25.56%) /assets/figures/2025-vllm-anatomy/fwd_pass.png -- 898.39kb -> 669.03kb (25.53%) /assets/figures/stack/stack-overview-2.png -- 435.36kb -> 326.14kb (25.09%) /assets/figures/struct-decode-intro/vllm-new-xgrammar.png -- 54.88kb -> 41.33kb (24.7%) /assets/figures/2025-vllm-anatomy/prefix_pt1.png -- 340.68kb -> 257.12kb (24.53%) /assets/figures/spec-decode/figure7.png -- 316.34kb -> 238.83kb (24.5%) /assets/figures/vllm-serving-amd/405b1.png -- 10.03kb -> 7.60kb (24.23%) /assets/figures/spec-decode/figure4.png -- 1,027.78kb -> 780.44kb (24.07%) /assets/figures/2025-11-10-bitwise-exact-rl/rounding-sequence.png -- 188.65kb -> 143.32kb (24.03%) /assets/figures/vllm-serving-amd/case02-num-scheduler-steps/Mean TTFT (ms).png -- 23.40kb -> 17.80kb (23.94%) /assets/figures/vllm-serving-amd/case01-chunked-prefill/Requests Per Second.png -- 25.50kb -> 19.44kb (23.76%) /assets/figures/vllm-serving-amd/70b1.png -- 10.44kb -> 7.97kb (23.63%) /assets/figures/spec-decode/figure1.png -- 961.60kb -> 734.73kb (23.59%) /assets/figures/2025-vllm-anatomy/specdec_pt2.png -- 203.24kb -> 155.69kb (23.4%) /assets/figures/vllm-serving-amd/case07-tensor-parallelism/Mean TPOT (ms).png -- 31.49kb -> 24.29kb (22.86%) /assets/figures/ptpc/PTPC121.png -- 26.69kb -> 20.80kb (22.09%) /assets/figures/struct-decode-intro/mermaid-intro.svg -- 15.28kb -> 11.98kb (21.61%) /assets/figures/deepseek-v3-2/mla-indexer-block.png -- 111.70kb -> 87.80kb (21.39%) /assets/figures/vllm-serving-amd/case04-max-seq-len-to-capture/Mean TTFT (ms).png -- 42.56kb -> 33.46kb (21.37%) /assets/figures/transformers-backend/transformers-backend.png -- 53.48kb -> 42.23kb (21.03%) /assets/figures/vllm-serving-amd/case08-max-num-seq/Request per Second.png -- 42.05kb -> 33.23kb (20.96%) /assets/figures/vllm-serving-amd/case06-kvcache-type/Mean TPOT (ms).png -- 33.58kb -> 26.57kb (20.88%) /assets/figures/2025-torch-compile/figure5_b.png -- 104.64kb -> 82.95kb (20.73%) /assets/figures/vllm-serving-amd/case01-chunked-prefill/Mean TTFT (ms).png -- 20.02kb -> 15.90kb (20.6%) /assets/figures/openrlhf-vllm/ray.png -- 106.83kb -> 84.96kb (20.48%) /assets/figures/vllm-serving-amd/405b2.png -- 13.67kb -> 10.87kb (20.47%) /assets/figures/vllm-serving-amd/case06-kvcache-type/Requests per Second.png -- 36.72kb -> 29.31kb (20.18%) /assets/figures/vllm-serving-amd/70b2.png -- 13.96kb -> 11.29kb (19.14%) /assets/figures/vllm-serving-amd/case07-tensor-parallelism/Requests per Second.png -- 32.98kb -> 26.71kb (19.02%) /assets/figures/semantic-router/request.png -- 104.31kb -> 84.53kb (18.97%) /assets/figures/aibrix/aibrix-diagram.png -- 209.78kb -> 170.00kb (18.96%) /assets/figures/qwen3-next/hybrid.png -- 24.10kb -> 19.55kb (18.86%) /assets/figures/vllm-serving-amd/case08-max-num-seq/Mean TPOT (ms).png -- 35.60kb -> 29.01kb (18.49%) /assets/figures/v1/v1_qwen2vl.png -- 177.75kb -> 145.00kb (18.43%) /assets/figures/vllm-tpu/whats-new.png -- 244.90kb -> 200.91kb (17.97%) /assets/figures/vllm-tpu/vllm-serve-model.png -- 75.67kb -> 62.20kb (17.8%) /assets/figures/perf-v060/overall_throughput.png -- 106.84kb -> 87.95kb (17.68%) /assets/figures/annimation0.gif -- 114.58kb -> 94.77kb (17.3%) /assets/figures/vllm-serving-amd/case07-tensor-parallelism/Mean TTFT (ms).png -- 28.37kb -> 23.73kb (16.37%) /assets/figures/semantic-router/signal.png -- 1,710.38kb -> 1,431.48kb (16.31%) /assets/figures/stack/stack-thumbnail.png -- 178.63kb -> 149.58kb (16.26%) /assets/figures/stack/stack-panel.png -- 204.38kb -> 171.16kb (16.25%) /assets/figures/agent-lightning/2_having.png -- 347.56kb -> 291.18kb (16.22%) /assets/figures/v1/v1_llama.png -- 275.17kb -> 230.64kb (16.18%) /assets/figures/llama31/perf_llama3.png -- 36.10kb -> 30.31kb (16.05%) /assets/figures/2025-vllm-sleep-mode/sleepmode.png -- 1,379.70kb -> 1,168.53kb (15.31%) /assets/figures/perf-v060/llama70B_comparison.png -- 93.75kb -> 79.57kb (15.13%) /assets/figures/perf-v060/throughput.png -- 228.48kb -> 194.82kb (14.73%) /assets/figures/vllm-serving-amd/case06-kvcache-type/Mean TTFT (ms).png -- 29.76kb -> 25.43kb (14.54%) /assets/figures/perf-v060/A100_70B.png -- 612.05kb -> 523.07kb (14.54%) /assets/figures/v1/v1_prefix_caching.png -- 124.52kb -> 106.58kb (14.41%) /assets/figures/perf-v060/llama8B_comparison.png -- 96.75kb -> 82.93kb (14.29%) /assets/figures/2025-torch-compile/figure1.png -- 31.63kb -> 27.22kb (13.95%) /assets/figures/vllm-serving-amd/case08-max-num-seq/Mean TTFT (ms).png -- 33.22kb -> 28.60kb (13.92%) /assets/figures/perf-v060/H100_70B.png -- 586.32kb -> 506.69kb (13.58%) /assets/figures/perf-v060/A100_8B.png -- 579.32kb -> 503.23kb (13.14%) /assets/figures/perf-v060/H100_8B.png -- 563.44kb -> 492.73kb (12.55%) /assets/figures/2025-vllm-on-intel-arc/perf-figure1.png -- 75.68kb -> 66.80kb (11.73%) /assets/figures/ptpc/PTPC-tumbnail.png -- 47.64kb -> 42.08kb (11.67%) /assets/figures/vllm-serving-amd/introduction/Throughput (Requests per Second).png -- 16.65kb -> 14.92kb (10.42%) /assets/figures/2025-vllm-on-intel-arc/perf-figure2.png -- 36.49kb -> 32.98kb (9.61%) /assets/figures/2025-vllm-on-intel-arc/perf-figure3.png -- 85.38kb -> 77.59kb (9.12%) /assets/figures/vllm-serving-amd/introduction/Mean TTFT (ms).png -- 22.26kb -> 20.50kb (7.91%) /assets/figures/distributed-inference/tp_strategies.png -- 84.22kb -> 77.62kb (7.84%) /assets/figures/vllm-2024-wrapped-2025-roadmap/vllm-contributor-groups.png -- 213.44kb -> 197.57kb (7.44%) /assets/figures/beyond-text/prithvi-prediction.png -- 4,898.67kb -> 4,550.72kb (7.1%) /assets/figures/2025-vllm-on-intel-arc/persistent-kernel2.png -- 50.61kb -> 47.51kb (6.12%) /assets/figures/semantic-router/full-params.png -- 1,609.11kb -> 1,513.74kb (5.93%) /assets/figures/2025-vllm-on-intel-arc/thread-load2.png -- 10.19kb -> 9.64kb (5.44%) /assets/figures/semantic-router/modular.png -- 1,793.29kb -> 1,697.53kb (5.34%) /assets/figures/semantic-router/lora.png -- 1,816.80kb -> 1,719.94kb (5.33%) /assets/figures/vllm-meetup/vllm_meetup_HSkim.png -- 3,673.41kb -> 3,485.31kb (5.12%) /assets/figures/2025-vllm-on-intel-arc/persistent-kernel1.png -- 48.81kb -> 46.38kb (4.98%) /assets/figures/lfai/vllm-lfai-light.png -- 416.27kb -> 396.23kb (4.81%) /assets/figures/2025-vllm-on-intel-arc/thread-load1.png -- 10.05kb -> 9.57kb (4.75%) /assets/figures/perf_a100_n1_dark.png -- 266.90kb -> 254.44kb (4.67%) /assets/figures/perf_a100_n3_dark.png -- 259.18kb -> 247.68kb (4.44%) /assets/figures/perf_a10g_n3_dark.png -- 254.55kb -> 243.36kb (4.4%) /assets/figures/perf_a10g_n1_dark.png -- 244.00kb -> 233.32kb (4.38%) /assets/figures/perf_a100_n1_light.png -- 284.77kb -> 272.66kb (4.25%) /assets/figures/perf_a10g_n1_light.png -- 260.24kb -> 249.59kb (4.09%) /assets/figures/perf_a10g_n3_light.png -- 271.77kb -> 261.02kb (3.96%) /assets/figures/vllm-meetup/vllm_meetup_Daniele.png -- 2,380.53kb -> 2,287.36kb (3.91%) /assets/figures/perf_a100_n3_light.png -- 275.74kb -> 265.05kb (3.88%) /assets/figures/vllm-meetup/vllm_meetup_nicolo.jpg -- 1,297.26kb -> 1,256.60kb (3.13%) /assets/figures/vllm-meetup/vllm_meetup_HJKim.jpg -- 1,113.18kb -> 1,088.45kb (2.22%) /assets/figures/vllm-meetup/image-3.png -- 2,606.86kb -> 2,578.41kb (1.09%) /assets/figures/vllm-meetup/image-6.png -- 3,119.93kb -> 3,097.48kb (0.72%) /assets/figures/minimax-m1/benchmark.png -- 208.27kb -> 207.39kb (0.42%) /assets/figures/ptpc/PTPCSpeedup.svg -- 10.01kb -> 9.97kb (0.39%) /assets/figures/2025-vllm-nvidia-nemotron/figure2.png -- 220.81kb -> 220.28kb (0.24%) /assets/figures/vllm-meetup/image-2.png -- 3,496.04kb -> 3,491.61kb (0.13%) /assets/figures/ptpc/PTPCReqs.svg -- 18.27kb -> 18.26kb (0.07%) /assets/figures/agent-lightning/4_tasks-spans-loop.svg -- 128.17kb -> 128.16kb (0.01%) Signed-off-by: ImgBotApp <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Beep boop. Your images are optimized!
Your image file size has been reduced by 18% 🎉
Details
📝 docs |
repo | 🙋🏾 issues | 🏪 marketplace
~Imgbot - Part of Optimole family