Models

Warm

llama3-8b-8k

meta-llama/Meta-Llama-3-8B

6,199

5,063,586

M
Warm

llama31-8b-16k

meta-llama/Meta-Llama-3.1-8B-Instruct

3,980

4,851,980

M
Warm

qwen25-7b-lc

Qwen/Qwen2.5-7B-Instruct

691

2,462,317

Q
Warm

mistral-v02-7b-std-lc

mistralai/Mistral-7B-Instruct-v0.2

2,795

1,813,026

M
Warm

qwen25-7b-lc

Qwen/Qwen2.5-7B-Instruct-1M

328

1,759,738

Q
Warm

llama31-8b-16k

deepseek-ai/DeepSeek-R1-Distill-Llama-8B

741

1,351,431

D
Warm

qwen3-8b

Qwen/Qwen3-8B

370

1,192,661

Q
Warm

llama31-8b-16k

meta-llama/Meta-Llama-3.1-8B

1,618

1,138,449

M
Warm

llama3-8b-8k

meta-llama/Meta-Llama-3-8B-Instruct

3,989

1,081,604

M
Warm

gemma3-27b

google/gemma-3-27b-it

1,099

964,080

G
Warm

llama31-70b-16k

meta-llama/Meta-Llama-3.1-70B-Instruct

810

945,390

M
Warm

qwen3-32b

Qwen/Qwen3-32B

381

930,441

Q
Warm

llama33-70b-16k

meta-llama/Llama-3.3-70B-Instruct

2,353

893,395

M
Warm

mistral-24b-lc

sarvamai/sarvam-m

233

861,089

S
Warm

qwen3-14b

Qwen/Qwen3-14B

175

752,144

Q
Warm

qwen25-7b-lc

deepseek-ai/DeepSeek-R1-Distill-Qwen-7B

653

572,177

D
Warm

qwen25-32b-lc

Qwen/Qwen2.5-32B-Instruct

276

557,442

Q
Warm

mistral-v02-7b-std-lc

HuggingFaceH4/zephyr-7b-beta

1,720

551,726

H
Warm

qwen25-14b-lc

Qwen/Qwen2.5-14B-Instruct

234

549,254

Q
Warm

qwen25-14b-lc

deepseek-ai/DeepSeek-R1-Distill-Qwen-14B

518

499,390

D