ModelPriceWatch.com
Last scan 2026-08-15 Models tracked 205 Providers 31 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

GLM-5.2 vs DeepSeek V4 Pro

Side-by-side comparison of API pricing, specs, benchmarks, and capabilities

DeepSeek V4 Pro is 74.7% cheaper on blended cost ($0.544 vs $2.15/Mtok) — note: different model classes (mid tier vs flagship)
Specification
by Z.AI
3 providers: Fireworks $1.40/$4.40 Together $1.40/$4.40 Z.AI $1.40/$4.40
4 providers: DeepSeek $0.435/$0.870 Fireworks $1.74/$3.48 Together $1.74/$3.48 DeepInfra $1.30/$2.60
Overview
StatusCurrent flagship Open weights Current mid tier
Released Jun 13, 2026 Apr 24, 2026
Pricing per million tokens
Input $1.40/Mtok $0.435/Mtok
Output $4.40/Mtok $0.870/Mtok
Blended avg $2.15/Mtok $0.544/Mtok
Cached input $0.260/Mtok $0.004/Mtok
Price basis Correct as published today. DeepSeek has announced that from 16:00 UTC on 2026-08-16 this model moves to peak/off-peak billing: peak $1.32 input / $0.044 cached input / $3.96 output per 1M tokens, off-peak exactly half ($0.66 / $0.022 / $1.98). Peak hours are 01:00-04:00 and 06:00-10:00 UTC; every other hour is off-peak. This row tracks the flat rate DeepSeek charges until that moment, and switches to the peak rate when it takes effect.
Specifications
Context window 1M tokens 1M tokens
Parameters Proprietary Proprietary
Speed (TPS)
Modalities
Input
text
text
Benchmarks source: Vellum LLM Leaderboard
Avg benchmark score 76 73.5
Perf / dollar 35.3 135.2
GPQA Diamond 91.2 90.1
SWE-Bench Verified 45 vendor 80.6 vendor
MCP Atlas 77 73.6
Humanity's Last Exam 54.7 48.2
ARC-AGI 2 30 vendor 30
AIME 2025 70 vendor 75
MMMLU 80 vendor 82
BFCL 70 vendor 72
HumanEval 84 vendor 80.6
MATH 500 76 vendor 80
Terminal-Bench 2.1 81
LiveCodeBench 93.5
BrowseComp 83.4
AIME 2026 99.2
Independent composite scores each on its own scale — not part of the average above
AA Intelligence Index 52.6 53.2
AA Agentic Index 45.7 49.6
AA-Omniscience Index (−100–100) 4.4 0.8
GDPval-AA v2 1506.1 ELO 1590 ELO
LMSYS Chatbot Arena 1469.4 ELO 1458.2 ELO
Providers
Available from
Fireworks — $1.40/$4.40/Mtok
Together — $1.40/$4.40/Mtok
Z.AI — $1.40/$4.40/Mtok
DeepSeek — $0.435/$0.870/Mtok
Fireworks — $1.74/$3.48/Mtok
Together — $1.74/$3.48/Mtok
DeepInfra — $1.30/$2.60/Mtok

Cost at scale

1M tokens · 50/50 input/output
Projected cost of GLM-5.2 vs DeepSeek V4 Pro at increasing token volumes
VolumeGLM-5.2DeepSeek V4 ProSavings
1M tokens $2.15 $0.54 $1.61 (74.9%)
10M tokens $21.5 $5.44 $16.06 (74.7%)
100M tokens $215 $54.37 $160.63 (74.7%)
1000M tokens $2150 $543.75 $1606.25 (74.7%)

When to pick which

distilled from the pricing and spec data above
Pick GLM-5.2 if…
  • raw capability matters more than price: 76 vs 73.5 average benchmark score
  • you want open weights — self-host it, fine-tune it, or exit the API entirely; DeepSeek V4 Pro is closed
Pick DeepSeek V4 Pro if…
  • cost dominates: $0.544/Mtok blended vs $2.15 — 74.7% less on the same 50/50 token mix
  • your workload re-reads context (agents, RAG, long chats): cached input costs $0.004/Mtok — 99% off its list input price
  • provider flexibility: available from 4 providers, so you can price-shop hosts or fail over
Try DeepSeek V4 Pro on DeepSeek

The cheaper option here — DeepSeek V4 Pro costs $0.544/Mtok blended on DeepSeek.

Get API key →

Summary

GLM-5.2 by Z.AI costs $1.40/Mtok input and $4.40/Mtok output, with a 1M-token context window. It supports text input and is available from 3 providers.

DeepSeek V4 Pro by DeepSeek costs $0.435/Mtok input and $0.870/Mtok output, with a 1M-token context window. It supports text input and is available from 4 providers.

On a blended cost basis, DeepSeek V4 Pro is 74.7% cheaper than GLM-5.2.

On benchmarks, GLM-5.2 scores higher (76 vs 73.5) on average. In terms of value, DeepSeek V4 Pro has better performance per dollar (135.2 vs 35.3).

Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.

More comparisons

pairs sharing a model with this page