Qwen3 Coder Next (Alibaba) vs Kimi K2.7 Code (Moonshot AI)
Qwen3 Coder Next (Alibaba) is 64% cheaper than Kimi K2.7 Code (Moonshot AI) ($1.80 vs $4.95 blended per 1M tokens).
Qwen3 Coder Next (Alibaba) on Alibaba vs Kimi K2.7 Code (Moonshot AI) on Moonshot AI — coding models. Every figure in the table below is the value shown on each model's own endpoint page, with its source and verification date. What the makers declare about the models themselves is further down.
| Metric | Qwen3 Coder Next (Alibaba) | Kimi K2.7 Code (Moonshot AI) |
|---|---|---|
| Input $/1M | $0.3 | $0.95 |
| Cached input $/1M | — | $0.19 |
| Output $/1M | $1.5 | $4 |
| Blended $/1M | $1.8 | $4.95 |
| Context window served | 256K | 256K |
| Max output served | 64K | — |
Highlighted cell = better on that row (lower price/latency, higher context/intelligence/throughput). "—" = not measured, or (cached input) no cache pricing published by that provider.
What it costs at your volume
| Monthly workload | Qwen3 Coder Next (Alibaba) | Kimi K2.7 Code (Moonshot AI) |
|---|---|---|
| Side project · 5M in / 1M out | $3 | $8.75 |
| Production app · 50M in / 10M out | $30 | $87.5 |
| Heavy pipeline · 500M in / 100M out | $300 | $875 |
Straight token math from the verified prices above — no caching, no batch discounts. Model your own traffic in the LLM cost calculator.
Kimi K2.7 Code (Moonshot AI)'s blended price has dropped 50% since 2026-09-03 ($9.9 → $4.95/1M). Full price history on each endpoint page.
Which to pick
Qwen3 Coder Next (Alibaba) if…
- cost per token is the constraint — $1.8 vs $4.95 blended per 1M
Kimi K2.7 Code (Moonshot AI) if…
…no measured axis on this page favors it — which is itself a data point. Rate limits, ecosystem and provider terms may still decide for it.
Every line above is derived from a measured row on this page — nothing editorial. What the table can't weigh: rate limits, SDK ecosystem, data-retention terms — those are on each endpoint page.
Same weights, other hosts
The table at the top compares one endpoint per side. But the same model is often sold by several hosts — open weights re-hosted, or a closed model resold through clouds — at different prices, sometimes a bigger spread than the head-to-head itself.
qwen3-coder-next is sold by 4 hosts we track. Cheapest: Parasail at $0.92/1M blended. All 4 priced side by side on the qwen3-coder-next page.
kimi-k2.7-code is sold by 15 hosts we track. Cheapest: StreamLake at $3.712/1M blended. All 15 priced side by side on the kimi-k2.7-code page.
What the makers declare
The table above compares two endpoints — what each host charges, how much of the window it serves, how fast it answers. This is the other register: what the maker declares about the weights, the same whichever host you buy from. Where a row here carries a different number from the same row above, both are true — one describes the model, the other describes the endpoint that serves it.
qwen3-coder-nextmaker-declared
- Context window
- 262,144 tokens
- Parameters
- 79,674,391,296≈ 79.7B
- Active experts
- 10 of 512 active per token
- Weight precision
- bfloat16
- License
- Apache 2.0 — open weights, permissivecurated · read 2026-09-07Scope: the licence on the published weights, and nothing else. It is not the maker's acceptable-use policy, which is a separate document, and it is not the contract you sign with whichever host you call — hosts set their own terms, which this index does not read.
- Knowledge cutoff
- not published by the maker≈ September 2025 third-party estimate — not the maker's figurecurated
- Modalities
- text
- Max output (official)
- to verify
- Release date
- to verify≈ February 3, 2026 third-party catalogue — not the maker's figure
- Maker lifecycle status
- to verify
Also declared in the weights
- Architecture class
- Qwen3NextForCausalLM
- Layers
- 48
- Vocabulary
- 151,936 tokens
- Weights first published
- January 30, 2026
- Weights last updated
- February 3, 2026
Qwen/Qwen3-Coder-Next · config.json @ a7fbcb5 · read 2026-10-08 · 6 of 10 fields published by the maker
kimi-k2.7-codemaker-declared
- Context window
- 262,144 tokens
- Parameters
- 1,026,879,376,368≈ 1.03T
- Active experts
- 8 of 384 active per token
- Weight precision
- not published by the maker
- License
- Modified MIT — open weights, conditionalcuratedScope: the licence on the published weights, and nothing else. It is not the maker's acceptable-use policy, which is a separate document, and it is not the contract you sign with whichever host you call — hosts set their own terms, which this index does not read.
- Knowledge cutoff
- not published by the maker≈ April 2026 third-party estimate — not the maker's figurecurated
- Modalities
- text · image · video
- Max output (official)
- to verify
- Release date
- to verify≈ June 12, 2026 third-party catalogue — not the maker's figure
- Maker lifecycle status
- to verify
Also declared in the weights
- Architecture class
- KimiK25ForConditionalGeneration
- Layers
- 61
- Vocabulary
- 163,840 tokens
- Weights first published
- June 11, 2026
- Weights last updated
- June 15, 2026
moonshotai/Kimi-K2.7-Code · config.json @ 74797c9 · read 2026-10-08 · 5 of 10 fields published by the maker
Sources: Qwen3 Coder Next (Alibaba) endpoint · Kimi K2.7 Code (Moonshot AI) endpoint · Qwen3 Coder Next (Alibaba) price verified 2026-10-09 · Kimi K2.7 Code (Moonshot AI) price verified 2026-10-23