The Known Good Updated 25 Jul 2026
DeepSeek: DeepSeek V3 vs Z.ai: GLM 4.7 ·DeepSeek vs Z.ai

DeepSeek: DeepSeek V3 vs Z.ai: GLM 4.7 — head-to-head comparison

DeepSeek: DeepSeek V3 profile Z.ai: GLM 4.7 profile All comparisons
DeepSeek: DeepSeek V3 — DeepSeek
Quality
#44 of 61
44based on 3 of 6 evaluations
Known Good Index
Speed
#280 of 316
20.9
Output tokens per second
Price
#105 of 340
$0.35
Blended per 1M tokens
In $0.20 · Out $0.80
Cache hit price
#10 of 206
$0.003
per 1M tokens (−99%)
Context
#225 of 345
164k
tokens (~300 pages)
Z.ai: GLM 4.7 — Z.ai
Quality
#26 of 61
68based on 3 of 6 evaluations
Known Good Index
Speed
#109 of 316
66.2
Output tokens per second
Price
#168 of 340
$0.74
Blended per 1M tokens
In $0.40 · Out $1.75
Cache hit price
#83 of 206
$0.080
per 1M tokens (−80%)
Context
#193 of 345
205k
tokens (~300 pages)
Summary

On the Known Good Index, DeepSeek: DeepSeek V3 scores 44 and Z.ai: GLM 4.7 scores 68 — a 24-point lead for Z.ai: GLM 4.7. They share 2 published evaluations, with DeepSeek: DeepSeek V3 ahead on 0 and Z.ai: GLM 4.7 ahead on 2.

Blended pricing is $0.35 per 1M tokens for DeepSeek: DeepSeek V3 and $0.74 for Z.ai: GLM 4.7 — DeepSeek: DeepSeek V3 is about 2.1× cheaper.

Throughput is 21 tokens per second for DeepSeek: DeepSeek V3 and 66 for Z.ai: GLM 4.7, with time to first token at 1.1s and 2.1s. DeepSeek: DeepSeek V3 accepts 164k tokens of context against 205k.

DeepSeek: DeepSeek V3 is open weights under Open weights (restricted use); Z.ai: GLM 4.7 is open weights under Open weights (unrestricted). 9 tracked hosts serve one or both, and 3 serve both.

Generated from tracked data on 25 Jul 2026 · how these comparisons are computed →

At a glance
MetricDeepSeek: DeepSeek V3Z.ai: GLM 4.7
Known Good Index 44 based on 3 of 6 evaluations 68 based on 3 of 6 evaluations
Blended $/1M $0.35 $0.74
Output tok/s 21 66
Context window 164k 205k
Providers 3 9
Jump to section

Quality

Known Good Index and the capability indices, side by side

IndicesComponents
Index scores

Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.

MetricDeepSeek: DeepSeek V3Z.ai: GLM 4.7
Known Good Index 44 68
Mathematics Index 40 83
Science Index 53 86
Reasoning Index 52 —
Agentic Index — 35
Openness Index 60 60
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = DeepSeek: DeepSeek V3 · marker = Z.ai: GLM 4.7 · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond57 / 83
MATH / AIME16 / 83

Right-hand figures read DeepSeek: DeepSeek V3 / Z.ai: GLM 4.7. DeepSeek: DeepSeek V3 leads on 0 of 2, Z.ai: GLM 4.7 on 2.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Arena ratings

Elo from blind pairwise votes, ingested from Arena. Higher is better.

MetricDeepSeek: DeepSeek V3Z.ai: GLM 4.7
Text Arena 1,333 1,436
Text Arena (style controlled) 1,358 1,442
The Known Good

Price & Cost

Published API pricing by token type

Per token typeCache discount
Published pricing

USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.

MetricDeepSeek: DeepSeek V3Z.ai: GLM 4.7
Input $/1M $0.20 $0.40
Output $/1M $0.80 $1.75
Blended $/1M $0.35 $0.74
Cache read $/1M $0.003 $0.080
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input accepted and maximum output emitted

Context and output limits

Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.

MetricDeepSeek: DeepSeek V3Z.ai: GLM 4.7
Context window 164k 205k
Max output tokens 16k 131k
The Known Good

Speed & Latency

Median throughput and time to first token across hosting providers

Output speedTime to first token
Throughput and latency

Medians across all tracked providers, derived from OpenRouter endpoint telemetry.

MetricDeepSeek: DeepSeek V3Z.ai: GLM 4.7
Output tokens/s 21 66
Time to first token 1.1s 2.1s
The Known Good

Providers

Which hosts serve each model, and at what price and speed

9 hosts · 3 serve both
Full provider benchmarks →
Provider DeepSeek: DeepSeek V3 blendedDeepSeek: DeepSeek V3 tok/s Z.ai: GLM 4.7 blendedZ.ai: GLM 4.7 tok/s
StreamLake $0.35 24 $0.80 41
DeepInfra $0.46 20 $0.74 49
Novita $0.62 14 $0.90 27
Cerebras — — $2.38 401
Google · first party — — $1.00 140
AtlasCloud — — $0.85 54
Venice — — $1.07 46
Phala — — $1.46 31
Z.AI · first party — — $1.00 30

Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.

The Known Good
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Elo — Arena Full attribution →