DeepSeek V4.1 Flash (max) vs GLM-5.3-Flash (max)
Results updated 31 October 2026 · Both in Claude Code · How we benchmark
Verdict
GLM-5.3-Flash (max) beats DeepSeek V4.1 Flash (max) by 4 points, and is cheaper and faster.
- Cost per task
- $0.20 vs $0.093
- Time per task
- 10.9 min vs 6.5 min (estimated)
- Tokens per attempt
- 1.58M vs 962k
- Image input
- Yes vs Yes
DeepSeek V4.1 Flash (max), Claude Code scores higher on GS-QA2; GLM-5.3-Flash (max), Claude Code scores higher on Earth-Bench V2 and GeoBenchX. GLM-5.3-Flash (max), Claude Code is cheaper and faster: $0.093 against $0.20 per task, and 6.5 min against 10.9 min.
Where they differ most
By kind of work, share of tasks passed, 0 to 100 · Higher is better
| Kind of work | DeepSeek V4.1 Flash (max) | GLM-5.3-Flash (max) |
|---|---|---|
| Spatial statistics | 31 | 78 |
| Recognising infeasible requests | 40 | 65 |
| GIS analysis | 92 | 79 |
| Making maps | 72 | 74 |
Where the tasks are set: each kind of work links to its tasks and scoresCharts for each: DeepSeek V4.1 Flash (max), Claude Code · GLM-5.3-Flash (max), Claude Code.
Side by side
| Metric | DeepSeek V4.1 Flash (max), Claude Code | GLM-5.3-Flash (max), Claude Code |
|---|---|---|
| Geospatial Agent Index: Text only | 68 | 72 |
| Geospatial Agent Index: Text + vision | No data | 73 |
| Earth-Bench V2 score | 58 | 69 |
| GeoBenchX score | 55 | 70 |
| GS-QA2 score | 89 | 76 |
| Vision agents score | No data | 78 |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 96 | 98 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 100 | 100 |
| OGD4All score (standalone; in no index) | 79 | 93 |
| GeoNatureAgent score (standalone; in no index) | 100 | 100 |
| MultiGlobeQA score (standalone; in no index) | 64 | 73 |
| GeoAgentEval score (standalone; in no index) | 84 | 96 |
| Remote sensing and raster analysis score | 75 | 74 |
| GIS analysis score | 92 | 79 |
| Climate and time series score | 64 | 67 |
| Spatial statistics score | 31 | 78 |
| Making maps score | 72 | 74 |
| Recognising infeasible requests score | 40 | 65 |
| Cost per task | $0.20 | $0.093 |
| Time per task (at 100 output tokens/s) | 10.9 min | 6.5 min (estimated) |
| Tokens per attempt | 1.58M | 962k |
| Turns per attempt | 26.0 | 18.9 |
| Input price per 1M tokens | $0.30 | $0.15 |
| Output price per 1M tokens | $1.20 | $0.50 |
| Context window | 1,048,576 | 1,048,576 |
| Image input | Yes | Yes |
| Developer | DeepSeek | Z.ai |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, DeepSeek V4.1 Flash (max), Claude Code or GLM-5.3-Flash (max), Claude Code?
GLM-5.3-Flash (max), Claude Code, 72 to 68.
Which is cheaper per task, DeepSeek V4.1 Flash (max), Claude Code or GLM-5.3-Flash (max), Claude Code?
GLM-5.3-Flash (max), Claude Code, $0.093 to $0.20.
Which writes fewer output tokens per task, DeepSeek V4.1 Flash (max), Claude Code or GLM-5.3-Flash (max), Claude Code?
GLM-5.3-Flash (max), Claude Code, 29k to 48k.
Which has the larger context window, DeepSeek V4.1 Flash (max), Claude Code or GLM-5.3-Flash (max), Claude Code?
They are level, at 1,048,576 tokens.