Created bymario-alka.dePowered bynoob2claw.detricoma.de
Open data · reproducible tests

Which model is fast & smart?

We benchmark AI language models on real hardware: speed in tokens per second and response time – plus quality on real agent tasks. Transparent, reproducible, directly comparable.

5036 benchmarks120 models3 test types
Live benchmark
Who generates text the fastest? Real measurements, real hardware.
1
NVIDIA-Nemotron-3.5-Lightning-30B-A3BGeForce RTX 3090
3.484,3tok/s
2
gemma-4-E2B-itGeForce RTX 5090
2.491,2tok/s
3
NVIDIA-Nemotron-3-Nano-4BRTX PRO 6000 Blackwe
2.182,7tok/s
4
gpt-oss-20bRTX PRO 6000 Blackwe
1.995,4tok/s
5
Ornith-1.0-35BRTX PRO 6000 Blackwe
1.799,4tok/s
Peak3.484tok/s
View ranking →
01

Real performance

TTFT, prefill and generation tokens/s instead of theoretical peak numbers.

02

Harness quality

Points and success rates from complete agent and chat runs.

03

Full transparency

Model, quantization, runtime and hardware stay visible on every run.

Community favorites

Top models by benchmarks

Speed (tokens/s) and quality (harness) of the most-tested models – directly comparable.

All models →
Current measurements

Leaderboard Snapshot

Alle Ergebnisse →
Metric:
Size
#Model / MakerMetricsParallelGPU / CPU / RAMRuntime
1NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
3.484,32 tok/s TG
Prefill 8.256 · TTFT 14.388 ms
10×NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAMtensorfoldgodclaw 1.1.6MLX 4-bit
2NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
2.867,15 tok/s TG
Prefill 8.107 · TTFT 17.054 ms
10×NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAMtensorfoldgodclaw 1.1.6MLX 4-bit
3gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
2.491,20 tok/s TG
Prefill 15.814 · TTFT 4.382 ms
10×NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAMllama.cppcodex_cliQ4_K_M
4gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
2.414,37 tok/s TG
Prefill 20.277 · TTFT 4.107 ms
10×NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAMllama.cppgodclawQ4_K_M
5gemma-4-E2B-it5BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation)
2.412,27 tok/s TG
Prefill 19.677 · TTFT 4.107 ms
10×NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAMllama.cppgodclawQ4_K_M
🎨 Bildgenerierung

Zuletzt erzeugte Bilder

Jede Kachel ist ein gemessener Lauf - Motiv, Modell und Generierungszeit auf echter Hardware.

Alle Bild-Benchmarks →
Hardware ranking

Benchmarks by graphics card

Generation throughput (tokens/s) per GPU – minimum, average and maximum across all performance runs.

All runs →
Hardware ranking

Benchmarks by CPU

Generation throughput (tokens/s) per CPU – minimum, average and maximum across all performance runs.

All runs →
AMD Ryzen 5 5600X 6-Core Processor202 Läufe
578,0tok/s Ø
Min 18,0Max 3.484,3
AMD Ryzen Threadripper PRO 9965WX 24-Cores519 Läufe
551,3tok/s Ø
Min 9,4Max 2.143,6
AMD Ryzen 9 9950X 16-Core Processor255 Läufe
503,7tok/s Ø
Min 0,0Max 2.414,4
AMD Ryzen 7 5800X3D 8-Core Processor221 Läufe
427,8tok/s Ø
Min 0,1Max 2.491,2
AMD Ryzen Threadripper PRO 3945WX 12-Cores6 Läufe
397,3tok/s Ø
Min 115,5Max 1.171,8
AMD Ryzen 9 8945HX with Radeon Graphics222 Läufe
259,3tok/s Ø
Min 0,0Max 1.492,8
AMD Ryzen 9 7945HX with Radeon Graphics99 Läufe
255,9tok/s Ø
Min 0,0Max 1.564,6
AMD Ryzen Threadripper PRO 3955WX 16-Cores189 Läufe
196,9tok/s Ø
Min 9,9Max 717,6
AMD Ryzen Threadripper PRO 7955WX 16-Cores1856 Läufe
165,8tok/s Ø
Min 0,0Max 1.568,0
AMD Ryzen 3 3100 4-Core Processor137 Läufe
162,9tok/s Ø
Min 21,9Max 561,4
NVIDIA Grace111 Läufe
144,5tok/s Ø
Min 11,4Max 903,3
AMD Zen2 8-Core (PlayStation 5)9 Läufe
139,4tok/s Ø
Min 10,6Max 302,6
AMD Ryzen 3 PRO 3200GE w/ Radeon Vega Graphics35 Läufe
134,1tok/s Ø
Min 35,1Max 361,8
AMD Ryzen Threadripper PRO 5975WX 32-Cores621 Läufe
132,6tok/s Ø
Min 0,7Max 1.744,0
AMD EPYC 7203P 8-Core Processor33 Läufe
111,0tok/s Ø
Min 9,0Max 388,1
Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz214 Läufe
109,0tok/s Ø
Min 0,1Max 609,5
NVIDIA RTX Spark N1X (6144-core GPU, 20-core CPU)20 Läufe
93,5tok/s Ø
Min 49,0Max 200,5
Intel(R) Core(TM) i5-7400 CPU @ 3.00GHz39 Läufe
68,6tok/s Ø
Min 6,8Max 133,7
AMD RYZEN AI MAX+ 395 w/ Radeon 8060S50 Läufe
64,1tok/s Ø
Min 7,0Max 326,9
Apple M6 (unified CPU/GPU SoC, ARM64, AdvSIMD)30 Läufe
56,8tok/s Ø
Min 9,2Max 298,1
Apple M6 (unified CPU/GPU SoC)30 Läufe
56,4tok/s Ø
Min 8,5Max 256,3
AMD Jaguar 8-Core (PlayStation 4)4 Läufe
13,0tok/s Ø
Min 1,3Max 43,2
4 x Cortex-A78C + 4 x Cortex-A5531 Läufe
5,5tok/s Ø
Min 0,0Max 19,6
Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz38 Läufe
4,0tok/s Ø
Min 0,0Max 14,5