Updated minutes ago
Claude 3 Opus
Anthropic · dense · 175B parameters · 200,000 context
Quality88.0
Architecture Details
TypeDENSE
Total Parameters175B
Active Parameters175B
Layers80
Hidden Dimension12,288
Attention Heads96
KV Heads16
Head Dimension128
Vocab Size152,064
Memory Requirements
BF16 Weights
350.0 GB
FP8 Weights
175.0 GB
INT4 Weights
87.5 GB
KV-Cache per Token327680 bytes
Activation Estimate4.50 GB
Fits on (single-node)
B200 SXM INT4B100 SXM INT4GB200 NVL72 (per GPU) INT4GB300 NVL72 (per GPU) INT4H200 SXM INT4H100 NVL 94GB (per GPU pair) INT4Instinct MI300X INT4Instinct MI325X FP8
GPU Recommendations
B200 NVL (pair)optimal
BF16 · 2 GPUs · tensorrt-llm
83/100
score
Throughput
280.0 tok/s
Cost/Month
$19929
Cost/M Tokens
$27.08
B200 SXMgood
BF16 · 4 GPUs · tensorrt-llm
78/100
score
Throughput
280.0 tok/s
Cost/Month
$17044
Cost/M Tokens
$23.16
H200 SXMgood
BF16 · 4 GPUs · tensorrt-llm
75/100
score
Throughput
280.0 tok/s
Cost/Month
$10211
Cost/M Tokens
$13.88
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| anthropic | $15.00 | $75.00 | Cheapest |
Quality Benchmarks
MMLU86.8
HumanEval67.0
GSM8K95.0
MT-Bench90.0
Capabilities
Features
✓ Tool Use✓ Vision✓ Code✓ Math✓ Reasoning✓ Multilingual✓ Structured Output
Supported Frameworks
Supported Precisions
BF16 (default)