Across 3 comparable benchmarks, GLM-4-9B-Chat has better reported results on 2 and worse on 1.
Moderate evidence3 of 3 shared results countedHow we compare
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | GLM-4-9B-Chat | DeepSeek-V3 |
|---|---|---|
| Context window | 128K tokens | 128K tokens |
| Input / 1M | — | $0.320 |
| Output / 1M | — | $0.890 |
| Blended (4:1) | — | $0.434 |
| Status | Available | Available |
| Released | June 5, 2024 | December 26, 2024 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 1 shared benchmark grouped by capability on a 0–100 scale. How averages are built
GLM-4-9B-Chat ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
GLM-4-9B-Chat ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Science & health, Computer & tool use, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show GLM-4-9B-Chat relative to DeepSeek-V3. Reading the results
| Benchmark | GLM-4-9B-Chat | DeepSeek-V3 | Gap |
|---|---|---|---|
| LiveCodeBenchAlgorithmic Coding · Pass @K | 51.8Normal · no tools | 34.6Normal · no tools | 1.50×#84 vs #101 of 113 |
| AIME 2024Mathematics · Accuracy | 76.4Normal · no tools | 39.0Normal · no tools | +37.4 pts#30 vs #47 of 54 |
| MMLU-ProKnowledge Exams · Accuracy | 72.4Normal · no tools | 75.9Normal · no tools | −3.5 pts#88 vs #80 of 153 |
Algorithmic Coding · Pass @K
Mathematics · Accuracy
Knowledge Exams · Accuracy
A dash means no result is available.
| Benchmark | GLM-4-9B-Chat | DeepSeek-V3 |
|---|---|---|
| FrontierMath | — | 1.7 |
| GPQA | 58.5 | — |
| GPQA Diamond | — | 59.1 |
| IMO-ProofBench Advanced | — | 4.3 |
| MATH | — | 87.8 |
| MATH-500 | — | 87.8 |
| MMLU | — | 88.5 |
| SciCode | — | 35.8 |
| Aider-Polyglot | — | 48.4 |
| HumanEval | — | 89.0 |
| Terminal Bench Hard | — | 6.8 |
| ECI | — | 132 |
| GDP.pdf | — | 0.8 |
| τ³-Banking | — | 4.7 |
| BBH | — | 92.3 |
| SimpleBench | — | 18.9 |
| SimpleQA | — | 24.9 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
GLM-4-9B-Chat: Listed text rates; pricing tier not confirmed.
DeepSeek-V3: Standard pricing. DeepInfra official models API (deepseek-ai/DeepSeek-V3) verified 2026-11-13
| Price / 1M tokens | GLM-4-9B-Chat | DeepSeek-V3 | Ratio |
|---|---|---|---|
| Input | — | $0.32 | — |
| Output | — | $0.89 | — |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | — | $0.434 | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
| Attribute | GLM-4-9B-Chat | DeepSeek-V3 |
|---|---|---|
| Availability | Available | Available |
| Provider | Zhipu AI | DeepSeek-AI |
| Model type | Chat model | Chat model |
| First release | June 5, 2024 | December 26, 2024 |
| Context window | 128K tokens | 128K tokens |
| Total parameters | 9B | 681B |
| Architecture | Dense | Dense |
| Documented capabilities | Multilingual | Multilingual |
| Input modalities | — | Text |
| Output modalities | — | Text |
| Weights | Open weights | Open weights |
| Licence | GLM-4 License | DEEPSEEK LICENSE AGREEMENT |
| Commercial use | Commercial use allowed | Commercial use allowed |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used GLM-4-9B-Chat or DeepSeek-V3? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.