ChatGLM2-6B

Released
2023-06-25
Last refreshed
2026-05-01
Status
Researched 180d ago
Open sourceCommercial use: permitted

ChatGLM2-6B has tracked benchmark evidence for general LLM work with open-source; evaluate it while provider pricing coverage matures.

Tsinghua Knowledge Engineering Group (THUDM) releases · 1 in the last 12 monthsChangelog →
Specifications
Family
ChatGLM2
Released
2023-06-25
Context
32K
Parameters
6.2B
Architecture
Decoder Only
Specialization
general
Openness
Open source
License
Apache 2.0OSI-approvedPermitted
Weights
Available
Code
Unknown
Training
Pretrained
Created by
Tsinghua Knowledge Engineering Group (THUDM)

Leading China's LLM innovation surge

Beijing, China
Founded 2018
Website
Pricing

No tracked provider pricing is available yet.

Links
Website HuggingFace

About

ChatGLM2-6B is the second-generation 6B open-source bilingual (Chinese-English) chat model from THUDM (Tsinghua University / Zhipu AI). Released June 25, 2023, it improved significantly over the original ChatGLM-6B: extended context to 32K tokens via FlashAttention, trained on 1.4T tokens (up from ~1T), and achieved approximately 23% better MMLU scores. Widely used as a Chinese-language LLM baseline. Succeeded by ChatGLM3-6B in October 2023.

Providers

No active provider routes are tracked for this model yet.

Capabilities

No model capability flags are currently sourced.

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Benchmark scores(2)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
GAOKAO
42.7zero-shot, objective-accuracyObserved 2023-05-20—
Source
LMArena Text Arena - Creative Writing
941.0text creative_writing 2026-10-22Observed 2026-10-22
Configuration: ChatGLM2 6B
Harness: LMArena blind pairwise human votes, Bradley-Terry
Evaluator: LMArena
Confidence: confirmed
Notes: LMArena Text arena, Creative Writing category (no style control), board published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @a467e6b9, text config, latest split). Board entry 'ChatGLM2 6B' (category creative_writing in text/latest-00000-of-00001.parquet): rank 409 of 416, rating 940.9 (95% CI 914.6-967.1), 566 votes. Source: LMArena (CC BY 4.0). Retrieved 2026-10-27.
Source

Source: LMArena (lmarena.ai), CC BY 4.0

Migration checks

No linked migration route is available for this model yet.

API versions

chatglm2-6b-32k