GLM 4.6
Native bilingual English/Chinese support for cross-lingual RAG without translation overhead. MIT license enables fine-tuning on proprietary knowledge bases with self-hosting via vLLM/SGLang.
Model Information
- Provider
- Zhipu AI
- License
- Open Source
- Input Price per 1M
- $0.40
- Output Price per 1M
- $1.75
- Context Window
- 203K
- Release Date
- 2025-09-30
- Model Name
- glm-4.6
- Total Evaluations
- 810
Performance Record
Wins346 (42.7%)
Losses337 (41.6%)
Ties127 (15.7%)
Wins
Losses
Ties
Performance Overview
ELO ratings by dataset
GLM 4.6's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.
GLM 4.6 - ELO by Dataset
Detailed Metrics
Dataset breakdown
Performance metrics across different benchmark datasets, including accuracy and latency percentiles.
SciFact
ELO 158032.2% WR87W-117L-66T
Quality Metrics
- Correctness
- 4.63
- Faithfulness
- 4.87
- Grounding
- 4.87
- Relevance
- 4.90
- Completeness
- 4.57
- Overall
- 4.77
Latency Distribution
- Mean
- 27880ms
- Min
- 3248ms
- Max
- 68513ms
MSMARCO
ELO 155144.1% WR119W-106L-45T
Quality Metrics
- Correctness
- 4.80
- Faithfulness
- 4.77
- Grounding
- 4.77
- Relevance
- 4.83
- Completeness
- 4.70
- Overall
- 4.77
Latency Distribution
- Mean
- 34694ms
- Min
- 9198ms
- Max
- 69527ms
PG
ELO 141251.8% WR140W-114L-16T
Quality Metrics
- Correctness
- 4.87
- Faithfulness
- 4.87
- Grounding
- 4.83
- Relevance
- 4.90
- Completeness
- 4.57
- Overall
- 4.81
Latency Distribution
- Mean
- 36774ms
- Min
- 9584ms
- Max
- 104257ms
Compare Models
See how it stacks up
Compare GLM 4.6 with other top llms to understand the differences in performance, accuracy, and latency.