Model
Model explorer
CodeGen2.5 7B
OPENSalesforce · CodeGen family · released Jul 6, 2023
7B code model matching 15B-16B peers (StarCoder-15B, CodeGen1/2-16B) via infill sampling and Flash-attention serving.
ReasoningCodingVisionFunction callingTool useAgentic
153.4
Elo · unrated
Parameters
7B
Active params
7B (dense)
Context
2K tokens
Architecture
Dense autoregressive transformer decoder (GPT-style, causal + infill objective)
License
Apache-2.0 (mono/multi variants; instruct variant is research-purposes-only)
Languages
—
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
best: Claude Opus 4.5 · 99.4%
Run it locally
VRAM @ Q4
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
GPTQ · GGUF
CodeGen family
Elo progression across releases
API price weights · each benchmark row carries its own source badge (see methodology)