DeepSeek-V4-Flash vs GLM-4.5-Air
Specs, VRAM requirements and download trends — updated 24 October 2026.
deepseek-ai · A · Open weights
Params
158.1 B
Context
1,048,576
30d
809 K
mit · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | DeepSeek-V4-Flash | GLM-4.5-Air |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 158.1 B | 110.5 B |
| Architecture | DeepseekV4ForCausalLM | Glm4MoeForCausalLM |
| Context length | 1,048,576 | 131,072 |
| Licence | mit | mit |
| Commercial use | Allowed | Allowed |
| Languages | — | 2 |
| Downloads 30d | 809,273 | 850,129 |
| Downloads all time | 12.9 M | 5.5 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-04-22 | 2025-07-20 |
| API input price / 1M tokens | $0.091 | $0.13 |
| API output price / 1M tokens | $0.18 | $0.85 |
| Est. cost per 1M input + 1M output tokens | $0.27 | $0.98 |
| Cached input / 1M tokens | $0.018 | $0.025 |
| AA Intelligence Index | 24.4 | 11.1 |
| AA Coding Index | 52 | — |
| AA Agentic Index | 26.3 | — |
| Max output tokens | 943,717 | 98,304 |
| API providers | 15 | 3 |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
DeepSeek-V4-Flash
$0.27
GLM-4.5-Air
$0.98
DeepSeek-V4-Flash is 3.6× cheaper at list price.
Download trend
Daily snapshots, last 89 days (28 Jul – 24 Oct)
3.1 M184 K
DeepSeek-V4-Flash
GLM-4.5-Air
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
GLM-4.5-Air · fp16 @ 128K ctx
508.7 GB / 24 GB
DeepSeek-V4-Flash · fp16 @ 1024K ctx
3,211.0 GB / 24 GB