Model
Model explorer

InternLM2.5-7B

OPEN
Shanghai AI Laboratory · InternLM2.5 family · released Jul 3, 2024

Widely used open 7B; near-perfect 1M-token needle retrieval via chat-1m checkpoint, state-of-the-art 7B-scale math reasoning.

ReasoningCodingVisionFunction callingTool useAgentic
738.6
Elo · rank #261
Parameters
7B
Active params
7B (dense)
Context
1M tokens
Architecture
Dense transformer decoder (GQA)
License
Apache-2.0
Languages
—
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
BIG-Bench HardReasoning71.6%#49
best: ERNIE 4.5 300B-A47B · 94.3%
CMMLUKnowledge78.0%#15
best: Doubao-1.5-Pro · 90.9%
GPQA DiamondReasoning38.4%#227
best: GPT-5.6 · 94.6%
GSM8KMath86.0%#62
best: Llama 3.1 405B · 96.8%
MATH-500Math60.1%#112
best: GPT-5 · 99.4%
MMLUKnowledge72.8%#132
best: OpenAI o3 · 92.9%
Run it locally
VRAM @ Q4
4.5 GB
VRAM @ FP16
14 GB
Fits on (Q4)
RTX 3060 12GBRTX 4070 Ti 16GBRTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB
Throughput data unavailable.
Quantizations
GGUF Q4
InternLM2.5 family
Elo progression across releases
InternLM2.5-7BJul 2024738.6
Resources
HFinternlm/internlm2_5-7b-chatGHInternLM/InternLMDOCSinternlm.readthedocs.io
API price weights · each benchmark row carries its own source badge (see methodology)