M
Manufacturer
Mistral AI
Franzoesisches KI-Labor; Mistral, Mixtral, Codestral, Devstral, Ministral, Magistral.
Best token generation
1.060,0 tok/s
Ministral-3-14B-Reasoning-2512
Best prefill
20.061,1 tok/s
Ministral-3-14B-Reasoning-2512
∅ Token generation
138,8 tok/s
Average of 728 runs
∅ Prefill
3.107,1 tok/s
Average of 728 runs
Best TTFT
20 ms
Magistral-Small-2509
Best harness rate
87%
Devstral-Small-2-24B-Instruct-2512
🏆
Ministral-3-14B-Reasoning-2512 · fastest model at 1.060,0 tok/s token generation
Best prefill of this manufacturer: 20.061,1 tok/s (Ministral-3-14B-Reasoning-2512)
Best prefill of this manufacturer: 20.061,1 tok/s (Ministral-3-14B-Reasoning-2512)
Top models by token generation
Best measured generation throughput per model (tok/s), published performance benchmarks.
Models
All benchmarks →11 assigned
Codestral-22B-v0.122B101 runs
128,1tok/s Ø
Min 8,9Max 641,4
Devstral-2-123B-Instruct-2512123B53 runs
20,9tok/s Ø
Min 0,0Max 143,1
Devstral-Small-2-24B-Instruct-251224B110 runs
152,7tok/s Ø
Min 8,9Max 725,6
Devstral-Small-250724B97 runs
149,4tok/s Ø
Min 9,7Max 647,6
Magistral-Small-250924B102 runs
171,7tok/s Ø
Min 4,0Max 737,0
Mamba-Codestral-7B-v0.17B1 runs
29,3tok/s Ø
Min 29,3Max 29,3
Ministral-3-14B-Reasoning-251214B126 runs
242,0tok/s Ø
Min 17,3Max 1.060,0
Mistral-Medium-3.5-128B128B31 runs
9,4tok/s Ø
Min 0,6Max 141,6
Mistral-Small-24B-Instruct-250123.6B5 runs
59,4tok/s Ø
Min 8,5Max 153,1
Mistral-Small-3.1-24B-Instruct-250324B80 runs
65,6tok/s Ø
Min 5,0Max 214,9
Mistral-Small-4-119B-2603119B26 runs
93,1tok/s Ø
Min 5,8Max 959,6

