MODEL PROOF
DeepSeek V3
BETTER VALUE OPTION
A stored alternative has an equal-or-higher measured score and an equal-or-lower price.
deal-only price: No offer passes the like-for-like test, so this price is a fallback: precision not disclosed. · precision not disclosed: The seller whose offer sets this price does not declare its serving precision. It is not read as full precision. · Serving precision differs between offers.
GLM 5.3 Flash · Δ score 137.1 points · 53% lower measured price · $0.238 / 1M tokens
GLM 4.5 Air is a further stored option with these losses: 164K → 131K context
- Independent LMArena score
- 1,332.5 ± 4.74 · 21,770 votes
- Context window
- 163.84K
- Maximum output
- 16K
- Input modalities
- text
- Output modalities
- text
- Published input price
- $0.286 / 1M tokens
- Published output price
- $1.14 / 1M tokens
- Pricing kind
- fixed
Inspect complete billing conditions and endpoint terms below.
Price from StreamLake: no serving offer passes the like-for-like test, so this is a fallback. deal-only price precision not disclosed The seller whose offer sets this price does not declare its serving precision. It is not read as full precision.
Cheaper right now: $0.257 in · $1.03 out per million tokens at StreamLake, a promotion of 10%. The price on this page is StreamLake’s standard rate: the ranking uses the standard rate, never a promotion.
Every price dimension · USD per million tokens
| Input | $0.286/M |
|---|---|
| Output | $1.14/M |
| Cached input | Unknown/M |
| Cache write | Unknown/M |
| Cache write 1h | Unknown/M |
| Reasoning | Unknown/M |
Who sells it
The same weights, different shops. Cheapest is not like-for-like when serving precision differs.
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.320/M from DeepInfra.
2 provider offers · rates, limits and conditions
Rates are USD per million tokens. Endpoint terms can differ even when the quoted price and precision match. A listed parameter is a provider declaration, not a task-success test.
| Seller | Input | Output | Precision | Uptime (1d) |
|---|---|---|---|---|
| StreamLakestreamlake Endpoint terms
| $0.257 | $1.03 | undeclared | 99.37% |
| DeepInfradeepinfra/fp4 Endpoint terms
| $0.320 | $0.890 | fp4 | 96.05% |
Independent scores
| LMArena Elo | 1332.5default |
|---|
LMArena Elo under CC BY 4.0. Absence of another board is not a score of zero.
Best rankings by task
- 3D #82 1095
- SVG #87 975
- UI components #89 1100
- Websites #91 1126
- Code categories #91 1112
- Data visualisation #95 1097
- Game development #98 1066
Per-task Elo and rank from Design Arena, via OpenRouter’s model feed. The rank is the feed’s, and it counts models OpenRouter does not list.
Capability
| Context window | 163K |
|---|---|
| Max output | 16K |
| Input modes | text |
| Tool use | yes |
| Reasoning | no |
| Knowledge cutoff | 2024-07-31 |
| Open weights | Unknown |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-10-06 |
| Quality data | verified |
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...
Questions this page answers
What does DeepSeek V3 cost?
$0.286/M in, $1.14/M out in this catalogue, as of the fetch date on this page. That is the API row, not a subscription.
Does DeepSeek V3 have an independent quality score?
DeepSeek V3 has an LMArena score in this catalogue.
What beats DeepSeek V3?
GLM 5.3 Flash has an equal-or-higher measured score and an equal-or-lower price.
Does this page use Artificial Analysis scores for DeepSeek V3?
No. Artificial Analysis figures are not published here. Quality on this page is LMArena Elo where a score exists; otherwise the row is unrated.
Is the cheapest DeepSeek V3 endpoint the same product?
Headline cheapest is a lower precision. Like-for-like at the best declared precision is $0.320/M from DeepInfra.
MODEL MONUMENT
Save or share this proof
Badge
The verdict as one SVG, for a README or a docs page. It states this model's dominance status at the balanced workload on the LMArena lens, with the date it was computed, and it is rebuilt with the catalogue — so it changes when the verdict changes, including to one you would rather it did not.
[](https://undominated.ai/models/deepseek__deepseek-chat/)<a href="https://undominated.ai/models/deepseek__deepseek-chat/"><img src="https://undominated.ai/badge/deepseek__deepseek-chat.svg" alt="DeepSeek V3 — Undominated.ai dominance verdict" height="36"></a>Direct file: https://undominated.ai/badge/deepseek__deepseek-chat.svg — a static SVG, written by scripts/build-badges.mjs on every build, so the copy you embed is never
older than the last deploy.
