Mid-range is the production spend band: expensive enough that you should measure it, cheap enough that it can carry daily work if the model fits.
Use this page as a routing guide, not a leaderboard. The right answer depends on the tool path, data policy, context size, and cost per successful task.
Current Shortlist
| Lane | Pricing anchor | Best use | Caveat |
|---|---|---|---|
| Claude Sonnet 5 | $2 / $10 through Aug 31; then $3 / $15 | First Claude cost/performance test | New tokenizer can produce about 30% more tokens for the same text |
| GLM-5.2 | $1.40 / $4.40 per 1M | July value pick to test for long-context supported-tool coding | AA and Z.AI benchmark signals still need repo-level checks |
| Kimi K3 API | $0.30 cache-hit / $3.00 cache-miss input, $15.00 output | Latest Kimi flagship and 1M-context eval lane | Open weights are pending until July 27; CAR is not verified |
| Kimi K2.7 Code API | $0.19 cache-hit / $0.95 cache-miss input, $4.00 output | Cheaper routine Kimi coding API lane | 256K context; thinking is required; HighSpeed costs more |
| MiniMax M3 Standard | $0.60 / $2.40 per 1M up to 512K input | Value coding-agent lane to test | Standard vs Priority pricing and Token Plan quota are separate |
| Kimi K2.6 API | $0.16 cache-hit / $0.95 cache-miss input, $4.00 output | Product/integration-specific compatibility lane | Keep only where non-thinking mode or an existing product contract still requires K2.6 |
| Gemini Flash/Pro lanes | model-specific | High-context Google workflow tests | Verify the exact Gemini model, tier, and data terms |
How To Choose
Start With The Tool Fit
If your team lives in Claude Code or claude.ai, start with Sonnet 5 and escalate to Opus only when Sonnet misses something important.
If your workflow supports GLM, MiniMax, Kimi, or Gemini through official or BYOK routes, run the same bug fix, refactor, and review task through those lanes before paying premium-model rates. GLM-5.2 is the July value pick to test because it combines 1M context, $1.40/$4.40 API pricing, AA Index 51, and supported-tool subscription routing. Kimi K3 is the new Kimi flagship to evaluate, but its $15 output price and pending weight release make CAR testing mandatory. Use GLM-5.2 vs Kimi K2.6/K2.7 for the cheaper Z.AI vs Kimi decision.
Measure Cost Per Success
Token price is only part of the bill. Track:
- prompt and output tokens,
- retries,
- failed patches,
- human review time,
- tests passed on first attempt,
- data policy and retention fit.
Escalate Deliberately
Move to Premium Models when the task proves it needs a stronger model:
- architecture or multi-file reasoning fails,
- long-horizon agent work loops,
- code review misses expensive defects,
- safety or compliance review needs a second pass,
- OpenAI or Claude tooling fit matters more than raw token price.
Production Routing Pattern
| If the task is… | Start with | Escalate to |
|---|---|---|
| Routine coding or summarization | Sonnet 5, GLM-5.2, or Kimi K2.7 Code | Kimi K3 or Opus 4.8 if quality matters |
| Supported-tool budget coding | GLM-5.2 | GPT-5.5 or Opus when GLM misses |
| Long-context agent work | MiniMax M3 or Gemini lane | Opus 4.8, then revisit Fable 5 only if access, compliance, and guardrail tests pass |
| OpenAI-native work | GPT-5.5 standard via the premium guide | GPT-5.5 Pro only after standard fails |
| Guardrail-sensitive cyber/bio/chem work | Approved access path first | Avoid Fable unless fallback/refusal is acceptable |
Sources
- Claude API docs: Pricing (Archive)
- OpenAI API docs: Pricing (Archive)
- Smart Spend Guide — current low-cost lane and subscription routing notes
- Premium LLM Comparison — Sonnet, Opus, Fable, and GPT-5.5 escalation ladder
- GLM-5.2 vs Kimi K2.6/K2.7 — current cheap coding-model comparison
- Claude Opus 4.8 — premium Claude baseline for escalation math
Related Resources
- /compare/models/ — Model comparison hub
- /compare/models/premium/ — Premium escalation guide
- /value/smart-spend/ — Current paid-stack strategy
- /tools/zai/ — GLM Coding Plan guide
- /models/claude-opus-4-8/ — Opus 4.8 baseline guide
- /models/claude-sonnet-5/ — Sonnet 5 launch, migration, and tokenizer guide
- /models/kimi-k2.7-code/ — Kimi K2.7 Code guide
- /models/kimi-k3/ — Kimi K3 launch guide
- /models/minimax-m3/ — MiniMax M3 guide
Last updated: July 18, 2026. Verify live provider pricing, availability, and open-weight status before committing large workloads.