zai-org

GLM-4.7-Flash

Name: GLM-4.7-Flash
Author: zai-org

31.2B parameters · mixture-of-experts · reasoning · Base · GLM family

Mixture-of-experts — what it means for memory & speed → Long context — how it uses memory →

mit hash verified source matched revision pinned

Good for: Long context

Which version should I download?

Default pick: GGUF · Q4_K_M — Standard 4-bit (Q4_K_M) accessible lead for GLM-4.7-Flash (31B-A? MoE, ~18 GB single file). MIT.

Set your rig to confirm it fits and to see if a higher-quality quant runs fully on your hardware.

Set your rig

Quantizations

Why these quants? Which quant should I pick? What do these mean?

Format	Level	Size	Verdict	Est. speed	Quality note	Swarm	Download
GGUF	IQ4_XS	15.15 GB	Set your rig	—	Low-bit 4-bit option (IQ4_XS) for GLM-4.7-Flash (31.221B MoE). MIT.	2S / 0L webseed OK	Magnet .torrent
GGUF	Q4_K_M	17.05 GB	Set your rig	—	Standard 4-bit (Q4_K_M) accessible lead for GLM-4.7-Flash (31B-A? MoE, ~18 GB single file). MIT.	2S / 0L webseed OK	Magnet .torrent
GGUF	Q5_K_M	19.94 GB	Set your rig	—	5-bit option (Q5_K_M) for GLM-4.7-Flash (31.221B MoE). MIT.	2S / 0L webseed OK	Magnet .torrent
GGUF	Q6_K	23.00 GB	Set your rig	—	6-bit option (Q6_K) for GLM-4.7-Flash (31.221B MoE). MIT.	2S / 0L webseed OK	Magnet .torrent
GGUF	Q8_0	29.66 GB	Set your rig	—	near-lossless 8-bit option (Q8_0) for GLM-4.7-Flash (31.221B MoE). MIT.	2S / 0L webseed OK	Magnet .torrent

GGUF · IQ4_XS

15.15 GB

Set your rig

Est. speed: —
Swarm: 2S / 0L webseed OK

Low-bit 4-bit option (IQ4_XS) for GLM-4.7-Flash (31.221B MoE). MIT.

Magnet .torrent

GGUF · Q4_K_M

17.05 GB

Set your rig

Est. speed: —
Swarm: 2S / 0L webseed OK

Standard 4-bit (Q4_K_M) accessible lead for GLM-4.7-Flash (31B-A? MoE, ~18 GB single file). MIT.

Magnet .torrent

GGUF · Q5_K_M

19.94 GB

Set your rig

Est. speed: —
Swarm: 2S / 0L webseed OK

5-bit option (Q5_K_M) for GLM-4.7-Flash (31.221B MoE). MIT.

Magnet .torrent

GGUF · Q6_K

23.00 GB

Set your rig

Est. speed: —
Swarm: 2S / 0L webseed OK

6-bit option (Q6_K) for GLM-4.7-Flash (31.221B MoE). MIT.

Magnet .torrent

GGUF · Q8_0

29.66 GB

Set your rig

Est. speed: —
Swarm: 2S / 0L webseed OK

near-lossless 8-bit option (Q8_0) for GLM-4.7-Flash (31.221B MoE). MIT.

Magnet .torrent

Want a different quant? Request it on the board →

Runs fully on

RTX 5090 · Q6_K RTX 4090 · Q5_K_M RTX 3090 · Q5_K_M RTX 3090 Ti · Q5_K_M RTX Pro 6000 Blackwell 96GB · Q8_0

Green = the model's best quant fits fully in GPU/unified memory at 8K context. Tap a card for its full "what runs on it" page.

Context

Advertised 198k · usable ≈ 198k (reported — maintainer docs)

How we know

Evidence grade: reported — a maintainer claim with no independent evaluation.

Z.AI docs: GLM-4.7 — observed Jul 18, 2026
GLM-4.7-Flash model card — observed Jul 18, 2026

Z.ai developer docs: context 200K, max output 128K. Card publishes capability benchmarks only; no long-context evals.

Reviewed on Jul 18, 2026.

Capabilities (as declared by the maintainer): reasoning tool calling

Run it

Runtime completeness (IQ4_XS torrent): llama.cpp ✅ Ollama – vision sidecar –

– Ollama — runs the GGUF directly; no Modelfile bundled
– vision sidecar — not a vision model

Quant

Context / KV 8,192 tokens · FP16 set above the quant table

Generic commands (no rig set). GPU-offload values assume the model fits on your GPU — set your rig for values tuned to your hardware.

Start the server

llama-server -m GLM-4.7-Flash-IQ4_XS.gguf -c 8192 -ngl 999

Use llama-cli in place of llama-server for a one-shot prompt. This model compresses its KV cache, but only on llama.cpp b5137 (April 2025) or newer AND a GGUF converted after that date. Older files -- including some still-popular DeepSeek-R1 and V3-0324 quants -- fall back to the uncompressed layout and use about 5 MB per token instead of 70 KB. Check the quant repo date if long context is the point.

Save as Modelfile next to the GGUF (Modelfile)

FROM ./GLM-4.7-Flash-IQ4_XS.gguf
PARAMETER num_ctx 8192
PARAMETER num_gpu 999
PARAMETER stop "<|endoftext|>"
PARAMETER stop "<|user|>"
PARAMETER stop "<|observation|>"
PARAMETER temperature 1
# Chat template: Ollama uses the template embedded in the GGUF (no TEMPLATE directive needed).

Create

ollama create zai-org-glm-4-7-flash -f Modelfile

Run

ollama run zai-org-glm-4-7-flash

GGUF is not a first-class vLLM format.

vLLM GGUF support is experimental and single-file only; prefer safetensors/GPTQ/AWQ for production. If you must, pass the .gguf path to `vllm serve` with --load-format gguf on a recent vLLM.

transformers does not load GGUF weights.

GGUF weights run under llama.cpp, Ollama or LM Studio.

Load GLM-4.7-Flash-IQ4_XS.gguf, set the context length to 8192 tokens. Set GPU offload to Max (all layers).

MLX runs MLX-format weights only (Apple Silicon). This quant is a GGUF build.

GGUF weights run under llama.cpp, Ollama or LM Studio.

Technical details

Chat template

[gMASK]<sop>
{%- if tools -%}
<|system|>
# Tools

You may call one or more functions to assist with the user query.

You are provided with function signatures within <tools></tools> XML tags:
<tools>
{% for tool in tools %}
{{ tool | tojson(ensure_ascii=False) }}
{% endfor %}
</tools>

For each function call, output the function name and arguments within the following XML format:
<tool_call>{function-name}<arg_key>{arg-key-1}</arg_key><arg_value>{arg-value-1}</arg_value><arg_key>{arg-key-2}</arg_key><arg_value>{arg-value-2}</arg_value>...</tool_call>{%- endif -%}
{%- macro visible_text(content) -%}
    {%- if content is string -%}
        {{- content }}
    {%- elif content is iterable and content is not mapping -%}
        {%- for item in content -%}
            {%- if item is mapping and item.type == 'text' -%}
                {{- item.text }}
            {%- elif item is string -%}
                {{- item }}
            {%- endif -%}
        {%- endfor -%}
    {%- else -%}
        {{- content }}
    {%- endif -%}
{%- endmacro -%}
{%- set ns = namespace(last_user_index=-1) %}
{%- for m in messages %}
    {%- if m.role == 'user' %}
        {% set ns.last_user_index = loop.index0 -%}
    {%- endif %}
{%- endfor %}
{% for m in messages %}
{%- if m.role == 'user' -%}<|user|>{{ visible_text(m.content) }}
{%- elif m.role == 'assistant' -%}
<|assistant|>
{%- set reasoning_content = '' %}
{%- set content = visible_text(m.content) %}
{%- if m.reasoning_content is string %}
    {%- set reasoning_content = m.reasoning_content %}
{%- else %}
    {%- if '</think>' in content %}
        {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
        {%- set content = content.split('</think>')[-1].lstrip('\n') %}
    {%- endif %}
{%- endif %}
{%- if ((clear_thinking is defined and not clear_thinking) or loop.index0 > ns.last_user_index) and reasoning_content -%}
{{ '<think>' + reasoning_content.strip() +  '</think>'}}
{%- else -%}
{{ '</think>' }}
{%- endif -%}
{%- if content.strip() -%}
{{ content.strip() }}
{%- endif -%}
{% if m.tool_calls %}
{% for tc in m.tool_calls %}
{%- if tc.function %}
    {%- set tc = tc.function %}
{%- endif %}
{{- '<tool_call>' + tc.name -}}
{% set _args = tc.arguments %}{% for k, v in _args.items() %}<arg_key>{{ k }}</arg_key><arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>{% endfor %}</tool_call>{% endfor %}
{% endif %}
{%- elif m.role == 'tool' -%}
{%- if m.content is string -%}
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
    {{- '<|observation|>' }}
{%- endif %}
{{- '<tool_response>' }}
{{- m.content }}
{{- '</tool_response>' }}
{%- else -%}
<|observation|>{% for tr in m.content %}
<tool_response>{{ tr.output if tr.output is defined else tr }}</tool_response>{% endfor -%}
{% endif -%}
{%- elif m.role == 'system' -%}
<|system|>{{ visible_text(m.content) }}
{%- endif -%}
{%- endfor -%}
{%- if add_generation_prompt -%}
    <|assistant|>{{- '</think>' if (enable_thinking is defined and not enable_thinking) else '<think>' -}}
{%- endif -%}

Sampling defaults

temperature: 1

Stop strings

<|endoftext|> <|user|> <|observation|>

Evidence & provenance

Source

Provenance: https://huggingface.co/zai-org/GLM-4.7-Flash
Revision pin: 7dd20894a642a0aa287e9827cb1a1f7f91386b67
Manifest: Present

License

Name: mit
Commercial use: yes
Access: Open

How verification works →

File hashes (SHA-256)

glm-4.7-flash-iq4-xs/GLM-4.7-Flash-IQ4_XS.gguf 44ad6ab1885f3fd2282b1b684df729868ef2da65fd8adeab4eb848b25ac3925b
glm-4.7-flash-iq4-xs/LICENSE 55719438a8c5280813899cd9afe7173f934bcafd60f90197eb8be44572b9bc0c
glm-4.7-flash-iq4-xs/README.md c3ca90d7a4c4969a151d58f48d3672d1a2f0b95d90a5856a125c169918730aec
glm-4.7-flash-q4-k-m/GLM-4.7-Flash-Q4_K_M.gguf 29837ed2c0fc5f51981adf8ac8083fcf80743c598381f13e9f06cbad0498b174
glm-4.7-flash-q4-k-m/LICENSE 55719438a8c5280813899cd9afe7173f934bcafd60f90197eb8be44572b9bc0c
glm-4.7-flash-q4-k-m/README.md c3ca90d7a4c4969a151d58f48d3672d1a2f0b95d90a5856a125c169918730aec
glm-4.7-flash-q5-k-m/GLM-4.7-Flash-Q5_K_M.gguf 167848c4b460887d7a52c44c378d3996a0d38b3302ac2045494dc8b50869a893
glm-4.7-flash-q5-k-m/LICENSE 55719438a8c5280813899cd9afe7173f934bcafd60f90197eb8be44572b9bc0c
glm-4.7-flash-q5-k-m/README.md c3ca90d7a4c4969a151d58f48d3672d1a2f0b95d90a5856a125c169918730aec
glm-4.7-flash-q6-k/GLM-4.7-Flash-Q6_K.gguf f5bdc431802d7a3dab3be5a50ad0848a5569fd45c98c5e2ad83412c87cd887dc
glm-4.7-flash-q6-k/LICENSE 55719438a8c5280813899cd9afe7173f934bcafd60f90197eb8be44572b9bc0c
glm-4.7-flash-q6-k/README.md c3ca90d7a4c4969a151d58f48d3672d1a2f0b95d90a5856a125c169918730aec
glm-4.7-flash-q8-0/GLM-4.7-Flash-Q8_0.gguf 8c0922b03326932a40315ce92bbee01cc836ad18664268019143c898bdd7dd0b
glm-4.7-flash-q8-0/LICENSE 55719438a8c5280813899cd9afe7173f934bcafd60f90197eb8be44572b9bc0c
glm-4.7-flash-q8-0/README.md c3ca90d7a4c4969a151d58f48d3672d1a2f0b95d90a5856a125c169918730aec

Explore further

Appears in

What runs on 24 GB of VRAM What runs on 32 GB of VRAM What runs on 48 GB of VRAM What runs on a 36 GB Mac What runs on a 48 GB Mac What runs on a 64 GB Mac

Also consider

Get an email when a better quant fits your rig.

Set your rig once; we'll alert you when a new or better-fitting build lands.

Set your rig

Performance reports

Real-world throughput reported by the community (and scraped sources).

View performance leaderboard

Community

Reviews, sampler presets and community runtime reports. Be the first to contribute.

Reviews (0)

No reviews yet.

Sampler presets

No presets yet.

Community runtime reports

Unofficial, community-submitted "it loads for me" reports. These are not the official verified-working badge — an admin reviews reports before anything is marked officially verified.

No community reports yet.

Recommended download

Quantizations

Runs fully on

Context

Run it

Chat template

Sampling defaults

Stop strings

Evidence & provenance

Source

License

File hashes (SHA-256)

Explore further

Appears in

Also consider

Performance reports

Reviews (0)

Sampler presets

Community runtime reports