Model comparison - Similar price, same tier
DeepSeek V4 Pro vs GLM-5.3
DeepSeek V4 Pro is from DeepSeek and GLM-5.3 is from Z.ai. On list price, DeepSeek V4 Pro charges 6% less than GLM-5.3 for input ($1.32 vs $1.40 per 1M tokens), and DeepSeek V4 Pro charges 10% less than GLM-5.3 for output ($3.96 vs $4.40 per 1M tokens). DeepSeek V4 Pro and GLM-5.3 both have a 1,000,000-token context window; max output is 393,216 and 128,000 tokens per response. DeepSeek V4 Pro was released on April 24, 2026 and GLM-5.3 on August 18, 2026. Both publish open weights.
Last verified from each provider's pricing page. All sources
Side-by-side specs
| Spec | DeepSeek V4 Pro | GLM-5.3 |
|---|---|---|
| Provider | DeepSeek | Z.ai |
| Input / 1M tokens | $1.32 | $1.40 |
| Output / 1M tokens | $3.96 | $4.40 |
| Cache read / 1M | $0.044 | $0.26 |
| Context window | 1,000,000 | 1,000,000 |
| Max output | 393,216 | 128,000 |
| Release date | ||
| Weights | Open | Open |
| License | MIT | GLM-5.3 license (Hugging Face license tag: other) |
| API model id | deepseek-v4-pro | glm-5.3 |
| AI Gateway id | deepseek/deepseek-v4-pro | zai/glm-5.3 |
| Last verified |
Prices in USD per 1M tokens at each provider's standard first-party rate. A dash means the provider does not bill that item separately or no primary page states it.
What each costs in practice
The same three workloads every model page uses, worked from each model's standard list rates. Swap in your own token counts with the math on each model's page.
| Workload | DeepSeek V4 Pro | GLM-5.3 |
|---|---|---|
| One agent session1M input tokens and 100K output tokens, no caching. | $1.72 | $1.84 |
| A chat app10,000 requests a day at 2K input and 500 output tokens each. | $46.20$1,386.00 per 30-day month | $50.00$1,500.00 per 30-day month |
| The same session at a 90% cache hit90% of the input is read from the prompt cache. The one-time cache write premium is not included. | $0.5676 | $0.814 |
Show the math
DeepSeek V4 Pro
- One agent sessionInput: 1M x $1.32 = $1.32Output: 0.1M x $3.96 = $0.396
- A chat appInput: 20M x $1.32 = $26.40Output: 5M x $3.96 = $19.80
- The same session at a 90% cache hitCached input: 0.9M x $0.044 = $0.0396Uncached input: 0.1M x $1.32 = $0.132Output: 0.1M x $3.96 = $0.396
GLM-5.3
- One agent sessionInput: 1M x $1.40 = $1.40Output: 0.1M x $4.40 = $0.44
- A chat appInput: 20M x $1.40 = $28.00Output: 5M x $4.40 = $22.00
- The same session at a 90% cache hitCached input: 0.9M x $0.26 = $0.234Uncached input: 0.1M x $1.40 = $0.14Output: 0.1M x $4.40 = $0.44
Benchmarks under the same setting
Only boards that list both models under the same run setting appear here, one row per board and setting, with each score exactly as the board shows it. Scores on different boards use different scales and are never combined.
| Board and setting | DeepSeek V4 Pro | GLM-5.3 |
|---|---|---|
| DeepSWE v1.1mini-swe-agent, max effort, Pass@1, v1.1 (113 tasks), best effort level shownOfficial leaderboard - deepswe.datacurve.ai | 63%±6% | 69%±3% |
Set them up in your tools
Step-by-step guides whose official docs list DeepSeek V4 Pro or GLM-5.3 by model id.
- Use DeepSeek in AiderLists DeepSeek V4 Pro
- Use DeepSeek in Claude CodeLists DeepSeek V4 Pro
- Use DeepSeek in ClineLists DeepSeek V4 Pro
- Use DeepSeek in Codex CLILists DeepSeek V4 Pro
- Use DeepSeek in ContinueLists DeepSeek V4 Pro
- Use DeepSeek in GitHub CopilotLists DeepSeek V4 Pro
- Use DeepSeek in GooseLists DeepSeek V4 Pro
- Use DeepSeek in Kilo CodeLists DeepSeek V4 Pro
- Use DeepSeek in OpenCodeLists DeepSeek V4 Pro
- Use DeepSeek in Roo CodeLists DeepSeek V4 Pro
- Use DeepSeek in ZedLists DeepSeek V4 Pro
- Use GLM in AiderLists GLM-5.3
- Use GLM in Claude CodeLists GLM-5.3
- Use GLM in ClineLists GLM-5.3
- Use GLM in Codex CLILists GLM-5.3
- Use GLM in ContinueLists GLM-5.3
- Use GLM in CursorLists GLM-5.3
- Use GLM in GooseLists GLM-5.3
- Use GLM in Kilo CodeLists GLM-5.3
- Use GLM in OpenCodeLists GLM-5.3
- Use GLM in Roo CodeLists GLM-5.3
- Use GLM in ZedLists GLM-5.3
Model pages
FAQ
How do DeepSeek V4 Pro and GLM-5.3 compare on price?
On list price, DeepSeek V4 Pro charges 6% less than GLM-5.3 for input ($1.32 vs $1.40 per 1M tokens), and DeepSeek V4 Pro charges 10% less than GLM-5.3 for output ($3.96 vs $4.40 per 1M tokens). For one agent session of 1M input and 100K output tokens with no caching, that is $1.72 on DeepSeek V4 Pro and $1.84 on GLM-5.3. Prices are each provider's first-party list rates, verified October 17, 2026 and October 17, 2026.
What are the context windows of DeepSeek V4 Pro and GLM-5.3?
DeepSeek V4 Pro and GLM-5.3 both have a 1,000,000-token context window; max output is 393,216 and 128,000 tokens per response.
Are DeepSeek V4 Pro and GLM-5.3 open weights?
DeepSeek V4 Pro's weights are published at huggingface.co under the MIT. GLM-5.3's weights are published at huggingface.co under the GLM-5.3 license (Hugging Face license tag: other).
More comparisons
- DeepSeek V4 Pro vs Gemini 4 ArgonOpen weights vs closed
- Gemini 4 Argon vs GLM-5.3Open weights vs closed
- DeepSeek V4 Pro vs Mistral Large 4Similar price, same tier
- GLM-5.3 vs Mistral Large 4Similar price, same tier
- DeepSeek V4 Pro vs GPT-6.1 SolOpen weights vs closed
- GLM-5.3 vs GPT-6.1 SolOpen weights vs closed
- Claude Sonnet 5.5 vs DeepSeek V4 ProOpen weights vs closed
- Claude Sonnet 5.5 vs GLM-5.3Open weights vs closed
Sources
DeepSeek V4 Pro verified from api-docs.deepseek.com; GLM-5.3 verified from docs.z.ai. This page is as current as the older of the two: .
- https://api-docs.deepseek.com/quick_start/pricing
- https://docs.z.ai/guides/overview/pricing
- https://api-docs.deepseek.com/updates
- https://api-docs.deepseek.com/news/news260424
- https://api-docs.deepseek.com/news/news260813
- https://api-docs.deepseek.com/news/news260910
- https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro
- https://vercel.com/ai-gateway/models/deepseek-v4-pro
- https://docs.z.ai/guides/llm/glm-5.3
- https://docs.z.ai/release-notes/new-released
- https://docs.z.ai/openapi.json
- https://huggingface.co/zai-org/GLM-5.3
- https://ai-gateway.vercel.sh/v1/models
Newsletter
Get model price changes as they ship
One email per week. New model releases, price cuts and what they mean for your bill.