GLM-4.7-Flash

Name: GLM-4.7-Flash
Price: 0.146 USD
Author: Z.ai (GLM)

Z.ai (GLM) · GLM-4.7 · released 2026-01-19 · data checked 2026-11-01

GLM-4.7-Flash is a 31B-parameter Mixture-of-Experts model (3B active per token) from Z.ai (GLM) (China), released 2026-01-19 under the MIT license. It scores 66.1 on the Small LM Index, #12 of 34 medium models. It needs about 20.4 GB of memory at 4-bit with an 8K-token context and costs $0.15 per million tokens through the cheapest API provider.

Small LM Index

66.1

#12 of 34 scored medium models

Likely 62.5–69.5 · from 5 of 6 core benchmarks

Parameters

31B

3B active · MoE

Price per 1M tokens

$0.15

$0.061 in / $0.40 out · blended 3:1

Context window

200K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best medium model on that benchmark. Sources: 6 reported by Z.ai (GLM). How we label sources.

GPQA Diamond Lab-reported Knowledge75.2

Class best: Nanbeige4.3 33B 89.5

Humanity's Last Exam Lab-reported Reasoning14.4

Class best: Qwen3.8 27B 30.8

AIME Lab-reported Math · AIME 202591.6

Class best: Ternary Bonsai 2 27B 95.8

LiveCodeBench Lab-reported Coding · LiveCodeBench v664.0

Class best: Qwen3.8 27B 90.3

SWE-bench Verified Lab-reported Coding59.2

Class best: Qwen3.6 27B 77.2

τ²-Bench Lab-reported Agentic79.5

Class best: MiniCPM-V 4.7 35B-A3B 84.7

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

MMLU-Pro≈ 82

Specifications

Lab: Z.ai (GLM) (CN · China)
Architecture: Mixture-of-Experts
Total parameters: 31B
Active parameters: 3B
Size class: Medium
Context window: 200K tokens
Input modalities: text
Reasoning: Reasoning (thinking)
License: MIT
Openness: Open weights
Memory to run: ~20.4 GB at 4-bit, 8K context (more sizes)
Hugging Face: 956.4K downloads (30 days) · 1.9K likes
API providers: 1 on OpenRouter
API price: $0.061 in / $0.40 out per 1M (OpenRouter, 2026-11-03)

Weights on Hugging Face ↗Try on OpenRouter ↗

Something wrong here? Report it.

Run it locally

Uses unsloth/GLM-4.7-Flash-GGUF (Q4_K_M). Swap the tag for another size: BF16, IQ4_NL, IQ4_XS, MXFP4, Q2_K, Q2_K_L, Q3_K_M, Q3_K_S….

Ollama

ollama run hf.co/unsloth/GLM-4.7-Flash-GGUF:Q4_K_M

llama.cpp

llama-server -hf unsloth/GLM-4.7-Flash-GGUF:Q4_K_M

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/GLM-4.7-Flash-4bit

vLLM (GPU server)

vllm serve zai-org/GLM-4.7-Flash

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context	200K context
16-bit F16	63 GB	65 GB	74 GB
8-bit Q8_0	34 GB	36 GB	45 GB
4-bit Q4_K_M	20 GB	22 GB	31 GB

Check it against your GPU, Mac or phone →

GLM-4.7-Flash

Benchmarks

Not reported: estimated

Specifications

Compare with

Ranked in

Sources

Run it locally