Model
Model explorer
Falcon-180B
OPENTII · Falcon family · released Sep 6, 2023
Billed as the world's most powerful openly released LLM at launch; trained on 3.5T RefinedWeb tokens, scored 67.85 avg on Open LLM Leaderboard.
ReasoningCodingVisionFunction callingTool useAgentic
402.5
Elo · rank #330
Parameters
180B
Active params
180B (dense)
Context
2K tokens
Architecture
Causal decoder-only Transformer, 180B (80 layers, multiquery attention, RoPE, FlashAttention)
License
TII Falcon-180B License and Acceptable Use Policy (commercial use permitted, excludes hosting use)
Languages
11+
API price (in/out)
No hosted API
Modalities
text
Benchmark results
Bar shows position within the tracked field; marker = field best
best: Hunyuan-T1 · 93.1
best: Llama 3.1 405B · 96.8%
best: PaLM 2 · 90.9%
Run it locally
VRAM @ Q4
100 GB
VRAM @ FP16
400 GB
Fits on (Q4)
M3 Ultra 512GBH200 141GBB200 192GB
Throughput data unavailable.
Quantizations
GGUF · GPTQ · AWQ
Falcon family
Elo progression across releases
API price weights · each benchmark row carries its own source badge (see methodology)