Catalog Changes
Version history starts when recording is enabled. Versions are created when reported values, array items or listing availability change. Array differences align positions, not inferred identities. The first observation is a baseline; unchanged downloads update last observation time.
Catalog Changes
-
swiss-ai/Apertus-8B-Instruct-2509 · 2026-10-13 12:01:11.397468 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 4998 1389 Providers[0] · Throughput 56.962025316455694 116.93319107972374 -
zai-org/GLM-4.7-Flash · 2026-10-13 12:01:11.385804 UTC
Current Model ListingHugging Face
4 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 1055.6666666666667 1090.6666666666667 Providers[0] · Throughput 77.85011226513758 68.74841313032933 Providers[2] · First Token Latency Ms 1503 1543 Providers[2] · Throughput 82.48783304462593 81.6926721673066 -
Qwen/Qwen3.5-27B · 2026-10-13 12:01:11.375086 UTC
Current Model ListingHugging Face
4 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 1458 1201 Providers[0] · Throughput 52.2365404682961 48.57785515307275 Providers[2] · First Token Latency Ms 274.6 231.8 Providers[2] · Throughput 55.57802709558149 49.64468028903892 -
moonshotai/Kimi-K3 · 2026-10-13 12:01:11.364025 UTC
Current Model ListingHugging Face
6 Changed Fields
Field / Array Position Previous Value Observed Value Providers[1] · First Token Latency Ms 2218.8 2450.5 Providers[1] · Throughput 29.654381494738782 26.962142801760685 Providers[3] · First Token Latency Ms 459.4 672.6 Providers[3] · Throughput 99.63680161309901 72.94535675322834 Providers[4] · First Token Latency Ms 2230 699 -
deepseek-ai/DeepSeek-V4-Pro-0813 · 2026-10-13 12:01:11.341470 UTC
Current Model ListingHugging Face
8 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 1490.3333333333333 1206.2 Providers[0] · Throughput 45.29939529754986 44.39461417638645 Providers[1] · First Token Latency Ms 489.4 698.8 Providers[1] · Throughput 72.25075911260151 115.20750083472944 Providers[2] · First Token Latency Ms 279.6 270.8 -
zai-org/GLM-5.1 · 2026-10-13 12:01:11.133532 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[2] · First Token Latency Ms 406.8 4675 Providers[2] · Throughput 72.95436812113138 49.857904970833125 -
meta-models/Muse-Glimmer-30B · 2026-10-13 12:01:11.121948 UTC
Current Model ListingHugging Face
4 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 583.6 298.4 Providers[0] · Throughput 98.80378995922686 103.32646388138728 Providers[2] · First Token Latency Ms 282.6 362.4 Providers[2] · Throughput 76.75745269137843 105.6275310850377 -
deepseek-ai/DeepSeek-V4-Flash · 2026-10-13 12:01:11.109818 UTC
Current Model ListingHugging Face
First recorded baseline; no earlier history is available.
-
deepseek-ai/DeepSeek-R1 · 2026-10-13 12:01:11.097216 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 913 1129 Providers[0] · Throughput 17.595101523735792 19.342750559945816 -
Qwen/Qwen2.5-VL-72B-Instruct · 2026-10-13 12:01:11.082551 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[1] · First Token Latency Ms 602.2 470.8 Providers[1] · Throughput 33.527977751319405 30.737422107012588 -
Qwen/Qwen3.5-9B · 2026-10-13 12:01:11.059657 UTC
Current Model ListingHugging Face
7 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 373.6 597.6 Providers[0] · Supports Structured Output True False Providers[0] · Throughput 97.93431053722091 62.92317657653598 Providers[2] · First Token Latency Ms 406.8 404.4 Providers[2] · Throughput 31.255161272416586 28.2167152686799 -
moonshotai/Kimi-K2.5 · 2026-10-13 12:01:11.047389 UTC
Current Model ListingHugging Face
5 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 954.6 1372 Providers[0] · Supports Structured Output False True Providers[0] · Throughput 72.37534628457715 46.7582613949488 Providers[2] · First Token Latency Ms 1726.2 1502 Providers[2] · Throughput 4.029864113302638 6.038319173474872 -
moonshotai/Kimi-K2.7-Code · 2026-10-13 12:01:11.035520 UTC
Current Model ListingHugging Face
4 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 1684 1778.4 Providers[0] · Throughput 44.58875060078775 25.571790175980276 Providers[2] · First Token Latency Ms 554.6 543.6 Providers[2] · Throughput 25.810804733563128 16.676284680905493 -
CohereLabs/command-a-plus-05-2026-bf16 · 2026-10-13 12:01:11.023146 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 258.4 3004 Providers[0] · Throughput 120.97261951470057 107.71721532649757 -
google/gemma-3-12b-it · 2026-10-13 12:01:11.010408 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[1] · First Token Latency Ms 248.6 225 Providers[1] · Throughput 38.71513226793361 54.89700850899103 -
zai-org/GLM-4.5-Air · 2026-10-13 12:01:10.995382 UTC
Current Model ListingHugging Face
4 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 722.6 891 Providers[0] · Throughput 79.22665098753257 74.22529030585508 Providers[1] · First Token Latency Ms 1840.8 1692.8 Providers[1] · Throughput 79.88580541930575 82.04240154833296 -
Qwen/Qwen3.5-397B-A17B · 2026-10-13 12:01:10.983047 UTC
Current Model ListingHugging Face
8 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 1410.4 1182 Providers[0] · Throughput 56.79860171164346 57.08128317570722 Providers[2] · First Token Latency Ms 638.4 363 Providers[2] · Throughput 81.37188077097605 94.26512997243375 Providers[3] · First Token Latency Ms 503.4 1759.4 -
Qwen/Qwen3-32B · 2026-10-13 12:01:10.972364 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[1] · First Token Latency Ms 479.4 219 Providers[1] · Throughput 26.20953325100975 24.18476247719466 -
zai-org/GLM-4.6 · 2026-10-13 12:01:10.960674 UTC
Current Model ListingHugging Face
4 Changed Fields
Field / Array Position Previous Value Observed Value Providers[0] · First Token Latency Ms 4311 3750.2 Providers[0] · Throughput 24.932067414749206 24.72099041957968 Providers[2] · First Token Latency Ms 506.4 422 Providers[2] · Throughput 11.689866881027948 74.99207542175265 -
Qwen/Qwen3-Coder-30B-A3B-Instruct · 2026-10-13 12:01:10.949028 UTC
Current Model ListingHugging Face
2 Changed Fields
Field / Array Position Previous Value Observed Value Providers[1] · First Token Latency Ms 346 391.4 Providers[1] · Throughput 130.26096251911804 145.60395573345414 - Next 20 Changes →