LocalFit

Catalog

Models 97

2026 releases ranked by the Artificial Analysis Intelligence Index (best evaluated mode). Arena ratings are a separate crowd signal, never blended. File sizes are actual Hugging Face bytes; “—” means no verified file.

Intelligence

Artificial Analysis Intelligence Index · top 15 ranked 2026 open-weight models · higher is better

  1. MiMo-V2.6-Pro-MOPD46.3
  2. GLM-5.344.8
  3. Step 5 Preview43.7
  4. Kimi K343.6
  5. GLM-5.3-Flash41.8
  6. Ling 3.1 Flash41.1
  7. Qwen3.8-2.4T-A95B39.9
  8. Qwen3.8-Flash-Next39.8
  9. DeepSeek V4.1 Flash39.5
  10. Muse Spark Open (816B-A42B)38.9
  11. MiMo-V2.6-Flash-MOPD37.9
  12. DeepSeek V4 Pro 081336.0
  13. Qwen3.8-27B33.7
  14. Motif 333.6
  15. K2-Horizon-375B-A23B30.5

AA estimate

ModelQualityGGUF file sizeMLX file size
#ModelReleasedSizeAA indexArenaQ4_K_MQ5_K_MQ8_04-bit8-bit
1MiMo-V2.6-Pro-MOPD
XiaomiMiMo/MiMo-V2.6-Pro-MOPD
2026-09-271.02T · 42B active46.31,491—————
2GLM-5.3
zai-org/GLM-5.3
2026-08-28MoE · size unknown44.81,471——746 GiB——
3Step 5 Preview
stepfun-ai/Step-5-Preview
2026-10 (day unknown)600B · 27B active43.71,441—————
4Kimi K3
moonshotai/Kimi-K3
2026-07-272.8T · 104B active43.61,475—————
5GLM-5.3-Flash
zai-org/GLM-5.3-Flash
2026-08-26320B · 18B active41.81,471——318 GiB190 GiB311 GiB
6Ling 3.1 Flash
inclusionAI/Ling-3.1-flash
2026 (month unknown)560B · 25B active41.1—333 GiB406 GiB567 GiB——
7Qwen3.8-2.4T-A95B
Qwen/Qwen3.8-2.4T-A95B
2026-08-122.4T · 95B active39.9———2,422 GiB——
8Qwen3.8-Flash-Next
Qwen/Qwen3.8-Flash-Next
2026-08-26180B · 6B active39.8———175 GiB104 GiB—
9DeepSeek V4.1 Flash
deepseek-ai/DeepSeek-V4.1-Flash
2026-09-10552B · 16B active39.51,463—————
10Muse Spark Open (816B-A42B)
meta-models/Muse-Spark-Open-816B-A42B
2026-10-15816B · 42B active38.9———812 GiB496 GiB—
11MiMo-V2.6-Flash-MOPD
XiaomiMiMo/MiMo-V2.6-Flash-MOPD
2026-09-27309B · 15B active37.91,455—————
12DeepSeek V4 Pro 0813
deepseek-ai/DeepSeek-V4-Pro-0813
2026-08-13MoE · size unknown36.01,446—————
13Qwen3.8-27B
Qwen/Qwen3.8-27B
2026-08-1427B dense33.71,44116.2 GiB19.5 GiB27.1 GiB15.0 GiB27.5 GiB
14Motif 3
Motif-Technologies/Motif-3
2026-08 (day unknown)314B · 13B active33.6 AA estimate—183 GiB220 GiB——312 GiB
15K2-Horizon-375B-A23B
IFM/K2-Horizon-375B-A23B
2026-09-03375B · 23B active30.5—221 GiB270 GiB375 GiB——
16Reflection Beam 1.0 501B-A23B
reflection/Beam-1.0-501B-A23B
2026-10-16501B · 23B active30.3———496 GiB——
17MiniMax-M3
MiniMaxAI/MiniMax-M3
2026-06 (day unknown)428B · 23B active29.21,433243 GiB284 GiB422 GiB225 GiB421 GiB
18Nex-N2-Pro (Qwen3.5-397B-A17B based)
nex-agi/Nex-N2-Pro
2026-06 (day unknown)397B · 17B active28.2 AA estimate—225 GiB264 GiB393 GiB—392 GiB
19NVIDIA Nemotron 3.5 Super VL 120B-A12B
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
2026-10-15120B · 12B active26.51,409——123 GiB64.6 GiB—
20Kimi K2.7 Code
moonshotai/Kimi-K2.7-Code
2026-06-121T · 32B active25.8—578 GiB——597 GiB—
21Inkling-Small
thinkingmachines/Inkling-Small
2026-07-30276B · 12B active25.71,413——261 GiB143 GiB—
22K2-Horizon-MoVA-36B-A4B
IFM/K2-Horizon-MoVA-36B-A4B
2026-09-0336B · 4B active25.3—20.8 GiB24.6 GiB37.1 GiB24.6 GiB37.1 GiB
23Hy3
tencent/Hy3
2026-07-06295B · 21B active25.31,438170 GiB————
24Inkling
thinkingmachines/Inkling
2026-07-15975B · 41B active25.01,441——798 GiB511 GiB—
25Solar Open 2 (250B-A15B)
upstage/Solar-Open2-250B
2026-07-22250B · 15B active24.7 AA estimate———248 GiB131 GiB248 GiB
26Ling-3.0-flash-VL
inclusionAI/Ling-3.0-flash-VL
2026-09-09124B · 5.5B active24.6—73.3 GiB89.3 GiB123 GiB——
27NVIDIA Nemotron 3 Ultra 550B-A55B
nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16
2026-06-04550B · 55B active22.91,445321 GiB369 GiB544 GiB288 GiB544 GiB
28Ling-3.0-flash-Fin
inclusionAI/Ling-3.0-flash-Fin
2026 (month unknown)124B · 5.1B active22.6—72.5 GiB84.4 GiB126 GiB——
29G9v3-39A5B (preview)
ai9stars/G9v3-39A5B
2026-08 (day unknown)39B · 5B active21.8 AA estimate—21.9 GiB—38.6 GiB21.2 GiB—
30A.X K2 (SK Telecom)
skt/A.X-K2
2026-07-29688B · 33B active21.5——————
31Qwen3.5-397B-A17B
Qwen/Qwen3.5-397B-A17B
2026-02-16397B · 17B active21.4 AA estimate1,438227 GiB273 GiB393 GiB208 GiB393 GiB
32NVIDIA Nemotron 3.5 Bolt 74B-A9B
nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16
2026-10-1574B · 9B active21.4———73.4 GiB40.3 GiB74.7 GiB
33K2-Horizon-7B
IFM/K2-Horizon-7B
2026-09-03size unknown20.6—5.2 GiB6.0 GiB8.9 GiB6.7 GiB8.9 GiB
34K-EXAONE 2.0 750B-A37B
LGAI-EXAONE/K-EXAONE-2.0-750B-A37B
2026-07-31750B · 37B active19.7 AA estimate——————
35LongCat-2.0
meituan-longcat/LongCat-2.0
2026-07 (day unknown)1.6T · 48B active19.1——————
36Qwen3.6-35B-A3B
Qwen/Qwen3.6-35B-A3B
2026-04-1635B · 3B active18.2—19.0 GiB23.0 GiB34.4 GiB19.0 GiB35.1 GiB
37Qwen3.5-122B-A10B
Qwen/Qwen3.5-122B-A10B
2026-02-24122B · 10B active17.7 AA estimate1,41771.3 GiB85.2 GiB121 GiB64.8 GiB122 GiB
38Muse Glimmer 30B
meta-models/Muse-Glimmer-30B
2026-08-1030B dense17.51,38915.6 GiB18.7 GiB27.6 GiB18.1 GiB31.1 GiB
39Gemma 4 26B A4B (it)
google/gemma-4-26B-A4B-it
2026-04-0225B · 3.8B active16.7 AA estimate1,43415.6 GiB—25.0 GiB14.3 GiB26.0 GiB
40Ring-2.6-1T
inclusionAI/Ring-2.6-1T
2026-05-151T · 63B active16.6——————
41K2-Horizon-3.7B
IFM/K2-Horizon-3.7B
2026-09-03size unknown15.6—2.9 GiB3.4 GiB5.0 GiB3.9 GiB5.0 GiB
42Gemma 4 31B (it)
google/gemma-4-31B-it
2026-04-0231B dense14.71,44317.1 GiB20.2 GiB30.4 GiB17.1 GiB31.4 GiB
43Mistral Medium 3.5 (128B dense)
mistralai/Mistral-Medium-3.5-128B
2026-04-28128B dense14.21,42169.8 GiB82.3 GiB124 GiB72.6 GiB—
44Gemma 4 12B (it)
google/gemma-4-12B-it
2026-06-0312B dense14.2 AA estimate—7.1 GiB8.2 GiB11.8 GiB6.3 GiB11.8 GiB
45Qwen3.5-9B
Qwen/Qwen3.5-9B
2026-03-029B dense13.3 AA estimate—5.3 GiB6.1 GiB8.9 GiB5.5 GiB9.7 GiB
46EXAONE 4.5 33B
LGAI-EXAONE/EXAONE-4.5-33B
2026-04-0933B dense13.2 AA estimate—18.7 GiB21.9 GiB32.7 GiB——
47Command A+ (05-2026)
CohereLabs/command-a-plus-05-2026-bf16
2026-05-20218B · 25B active13.1—123 GiB144 GiB216 GiB——
48Qwen3.5-4B
Qwen/Qwen3.5-4B
2026-03-024B dense13.1 AA estimate—2.6 GiB2.9 GiB4.2 GiB2.8 GiB4.8 GiB
49NVIDIA Nemotron 3.5 Lightning 30B-A3B
nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
2026-08-1130B · 3B active12.91,32923.7 GiB25.1 GiB32.6 GiB16.6 GiB31.3 GiB
50IBM Granite 4.2 30B
ibm-granite/granite-4.2-30b
2026-08-2530B dense12.71,36116.5 GiB19.3 GiB29.0 GiB15.3 GiB29.0 GiB
51MiniCPM5-2B
openbmb/MiniCPM5-2B
2026-09-072.5B dense12.5—1.5 GiB1.8 GiB2.5 GiB1.3 GiB2.5 GiB
52HyperNova 60B 2605 (compressed gpt-oss-120b)
MultiverseComputingCAI/Hypernova-60B-2605
2026-05-0660B · 4.8B active11.7 AA estimate—37.9 GiB41.3 GiB58.1 GiB——
53Nemotron Cascade 2 30B A3B
nvidia/Nemotron-Cascade-2-30B-A3B
2026-03-1930B · 3B active11.7 AA estimate—22.8 GiB24.3 GiB31.3 GiB16.6 GiB31.3 GiB
54LongCat-Flash-Lite
meituan-longcat/LongCat-Flash-Lite
2026-01 (day unknown)69B · 3B active11.5 AA estimate—37.5 GiB44.7 GiB67.9 GiB35.9 GiB67.8 GiB
55Mistral Small 4 (119B-2603)
mistralai/Mistral-Small-4-119B-2603
2026-03-16119B · 6B active11.3—67.6 GiB79.2 GiB118 GiB——
56Granite 4.2 8B
ibm-granite/granite-4.2-8b
2026-08-258B dense11.11,3205.0 GiB5.8 GiB8.7 GiB4.6 GiB8.7 GiB
57Ling-3.0-tiny
inclusionAI/Ling-3.0-tiny
2026-08 (day unknown)7.9B · 1.3B active11.1—4.5 GiB5.2 GiB7.8 GiB4.1 GiB—
58G9v3-3B
AI9Stars/G9v3-3B
2026-07 (day unknown)3B dense10.8 AA estimate—1.7 GiB2.0 GiB3.0 GiB1.6 GiB—
59Trinity-Large-Thinking
arcee-ai/Trinity-Large-Thinking
2026-04-01398B · 13B active10.8—225 GiB264 GiB395 GiB209 GiB394 GiB
60Nemotron 3 Nano Omni 30B A3B Reasoning
nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16
2026-04-2830B · 3B active10.3 AA estimate—22.8 GiB—31.3 GiB18.3 GiB33.3 GiB
61North Mini Code
CohereLabs/North-Mini-Code-1.0
2026-06-0930B · 3B active9.9—17.5 GiB20.3 GiB30.2 GiB17.2 GiB31.0 GiB
62DiffusionGemma 26B A4B
google/diffusiongemma-26B-A4B-it
2026-06-1025B · 3.8B active9.5 AA estimate—15.7 GiB17.8 GiB25.0 GiB15.4 GiB26.1 GiB
63Qwen3-Coder-Next
Qwen/Qwen3-Coder-Next
2026-02-0280B · 3B active9.2—45.2 GiB52.9 GiB79.0 GiB41.8 GiB78.8 GiB
64Granite 4.2 3B
ibm-granite/granite-4.2-3b
2026-08-253B dense9.11,2932.1 GiB2.4 GiB3.6 GiB1.9 GiB3.6 GiB
65Tri-21B-Think
trillionlabs/Tri-21B-Think
2026-02 (day unknown)21B dense9.0 AA estimate—11.7 GiB13.7 GiB20.5 GiB——
66Gemma 4 E4B (it)
google/gemma-4-E4B-it
2026-04-028B dense8.9—4.6 GiB5.1 GiB7.6 GiB4.8 GiB8.3 GiB
67MiniCPM5-1B
openbmb/MiniCPM5-1B
2026-05-191.1B dense8.8 AA estimate—0.6 GiB0.7 GiB1.1 GiB0.6 GiB1.1 GiB
68Sarvam 105B
sarvamai/sarvam-105b
2026-03-06105B · 10B active8.8 AA estimate—59.8 GiB—105 GiB——
69Nanbeige4.1-3B
Nanbeige/Nanbeige4.1-3B
2026-02 (day unknown)3B dense8.4 AA estimate—2.3 GiB2.6 GiB3.9 GiB2.1 GiB3.9 GiB
70LFM2.5-2.6B
LiquidAI/LFM2.5-2.6B
2026-08-042.7B dense8.4 AA estimate—1.6 GiB1.8 GiB2.7 GiB1.5 GiB2.7 GiB
71Falcon-H1R-7B
tiiuae/Falcon-H1R-7B
2026-01-057B dense7.8 AA estimate—4.3 GiB5.0 GiB7.5 GiB4.0 GiB7.5 GiB
72Gemma 4 E2B
google/gemma-4-E2B-it
2026-04-025.1B dense7.8 AA estimate—2.9 GiB3.1 GiB4.7 GiB3.3 GiB5.5 GiB
73Step3 VL 10B
stepfun-ai/Step3-VL-10B
2026-01 (day unknown)10B dense7.7 AA estimate—4.7 GiB5.4 GiB8.1 GiB——
74LFM2.5-8B-A1B
LiquidAI/LFM2.5-8B-A1B
2026-05-288.3B · 1.5B active7.2 AA estimate—4.8 GiB5.6 GiB8.4 GiB4.5 GiB8.4 GiB
75NVIDIA Nemotron 3 Nano 4B
nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16
2026-03-164B dense4.8—2.6 GiB2.9 GiB3.9 GiB2.1 GiB3.9 GiB
—LFM2.5-24B-A2B
LiquidAI/LFM2.5-24B-A2B
2026-10-1324B · 2B active——13.5 GiB15.9 GiB23.7 GiB12.6 GiB23.7 GiB
—MiniCPM-V 4.7 35B-A3B
openbmb/MiniCPM-V-4.7-35B-A3B
2026-10-1235B · 3B active——19.7 GiB23.0 GiB34.4 GiB21.9 GiB37.1 GiB
—Mellum2.1-12B-A2.5B-Thinking
JetBrains/Mellum2.1-12B-A2.5B-Thinking
2026-10-0812B · 2.5B active——7.5 GiB9.3 GiB12.0 GiB——
—Aleph Alpha Kolibri-1
Aleph-Alpha/Kolibri-1
2026-10-0378B · 3.5B active——44.2 GiB51.8 GiB77.4 GiB41.3 GiB—
—MiMo-V2.6-Distill-Qwen-9B
XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B
2026-09-229B dense——5.4 GiB6.4 GiB8.9 GiB5.5 GiB9.7 GiB
—K2-Horizon-0.9B
IFM/K2-Horizon-0.9B
2026-09-03size unknown——0.6 GiB0.7 GiB1.1 GiB0.8 GiB1.1 GiB
—Hy4-preview
tencent/Hy4-preview
2026-08-28770B · 49B active——435 GiB——404 GiB—
—Ling-3.0-flash
inclusionAI/Ling-3.0-flash
2026-08-07124B · 5.1B active——72.5 GiB84.4 GiB126 GiB65.2 GiB123 GiB
—Ornith-1.5-9B
ornith-ai/Ornith-1.5-9B
2026-08 (day unknown)9B dense——5.4 GiB6.2 GiB9.1 GiB4.7 GiB8.9 GiB
—DeepSeek V4 Flash 0731
deepseek-ai/DeepSeek-V4-Flash-0731
2026-07-31MoE · size unknown—————144 GiB—
—Poolside Laguna S 2.1
poolside/Laguna-S-2.1
2026-07-21118B · 8B active——89.4 GiB—120 GiB61.6 GiB116 GiB
—GLM-5.2
zai-org/GLM-5.2
2026-06-16MoE · size unknown————746 GiB390 GiB736 GiB
—Step 3.7 Flash
stepfun-ai/Step-3.7-Flash
2026-05-29198B · 11B active————195 GiB103 GiB195 GiB
—Qwen3.6-27B
Qwen/Qwen3.6-27B
2026-04-2227B dense——15.7 GiB18.2 GiB26.6 GiB15.0 GiB27.5 GiB
—Kimi K2.6
moonshotai/Kimi-K2.6
2026-04-201T · 32B active—————612 GiB—
—NVIDIA Nemotron 3 Super 120B-A12B
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16
2026-03-11120B · 12B active————120 GiB63.3 GiB119 GiB
—Qwen3.5-0.8B
Qwen/Qwen3.5-0.8B
2026-03-020.8B dense——0.5 GiB0.5 GiB0.8 GiB0.6 GiB0.9 GiB
—Qwen3.5-2B
Qwen/Qwen3.5-2B
2026-03-022B dense——1.2 GiB1.3 GiB1.9 GiB1.6 GiB2.5 GiB
—Qwen3.5-27B
Qwen/Qwen3.5-27B
2026-02-2427B dense——15.6 GiB18.3 GiB26.6 GiB15.0 GiB27.5 GiB
—Qwen3.5-35B-A3B
Qwen/Qwen3.5-35B-A3B
2026-02-2435B · 3B active——20.5 GiB24.4 GiB34.4 GiB19.0 GiB35.1 GiB
—GLM-4.7-Flash
zai-org/GLM-4.7-Flash
2026-01 (day unknown)30B · 3B active——17.1 GiB19.9 GiB29.7 GiB15.7 GiB29.6 GiB
—MiniMax-M2.7
MiniMaxAI/MiniMax-M2.7
2026 (month unknown)MoE · size unknown——129 GiB151 GiB226 GiB120 GiB226 GiB

Click a column to sort; unknown values sort last. AA = Artificial Analysis Intelligence Index of the evaluated hosted variant.